Electrophilic selenocyclization in 2-ene-1,5-diol systems: Unexpected oxetane vs. tetrahydrofuran formation

Article abstract of DOI:10.1016/S0040-4020(03)01134-7

Electrophile-induced cyclization of (E)- and (Z)-2-ene-1,5-diols to tetrahydrofurans and oxetanes is described. Significant differences between the present report and previous work have been noted. A tentative model is proposed.

Full text of DOI:10.1016/S0040-4020(03)01134-7

Tetrahedron 59 (2003) 7365–7376
Electrophilic selenocyclization in 2-ene-1,5-diol systems:
unexpected oxetane vs. tetrahydrofuran formation
*
´
Pierre Van de Weghe, Stephane Bourg and Jacques Eustache
´ ´
Laboratoire de Chimie Organique et Bioorganique associe au CNRS, Universite de Haute-Alsace,
´
Ecole Nationale Superieure de Chimie de Mulhouse 3, rue Alfred Werner, F-68093 Mulhouse Cedex, France
Received 11 April 2003; revised 3 July 2003; accepted 22 July 2003
Abstract—Electrophile-induced cyclization of (E)- and (Z)-2-ene-1,5-diols to tetrahydrofurans and oxetanes is described. Significant
differences between the present report and previous work have been noted. A tentative model is proposed.
q 2003 Elsevier Ltd. All rights reserved.
1. Introduction
and 4-exo modes compete.4 However, although the former
process (leading to tetrahydrofurans) is in principle
disfavoured,5 in almost all reported studies, using (Z)- or
(E)-1,2-disubstituted olefinic double bonds, the tetrahydro-
furan is the only product of the reaction,6 while we are
aware of only one instance where oxetane formation was
observed.7 This apparent deviation from Baldwin’s rules
has been explained by considering that the transition state in
such reactions is pyramidalized, therefore not correspond-
ing to a ‘true’ 5-endo-trig process.6d,e Fewer examples deal
with allylic alcohols and we could find only one example of
iodoetherification in a 2-ene-1,5-dihydroxy system related
to ours. In this case, formation of a single tetrahydrofuran
was also observed.8 Representative examples of such
studies are shown in Figure 3. Although no example exactly
matching our case had been published, the abundant data
available in related systems led us to be confident in the
validity of our approach.
The allylic/homoallylic system found in 2-ene-1,5-diols
offers many possibilities for chemical conversion. For
example, we recently described the use of (Z)-2-ene-1,5-
diols as key intermediates in the synthesis of spiroketals.1
Another synthetically useful possibility would be their
conversion into tetrahydrofurans (Fig. 1). Combined with
the mildness of the silicon-tethered RCM approach, which
allows the facile preparation of highly functionalized allylic
homoallylic diols, this conversion potentially provides
a new entry toward complex tetrahydrofurans as found in
e.g. acetogenins.
To test the feasibility of this approach, we chose
tetrahydrofuran A, a potential intermediate in acetogenin
synthesis,2 as a target molecule. The planned conversion
implies an electrophile-induced cyclization, following a 5-
endo mode, to afford a tetrahydrofuran, as shown in Figure 2.
Besides this desired process, however, two other possi-
bilities exist, leading to the formation of oxetanes or
oxiranes. The latter (3-exo cyclization), in turn, could suffer
nucleophilic opening by the homoallylic hydroxyl leading
again to the desired tetrahydrofurans. Thus, several
competing pathways leading to different products poten-
tially exist in 2-ene-1,5-diols making difficult to predict the
overall regio- and stereochemical outcome of the reaction.
2. Results and discussion
For the preparation of the required allylic homoallylic
precursor 7 (Scheme 1), we used the silicon-tethered RCM
approach recently developed by us1 and others (Scheme 1).9,10
Allylic alcohol 3 was prepared in 45% overall yield from
tridecanal by treatment with Ph3PvCH–COOMe,
reduction of the resulting a,b-unsaturated ester to the
corresponding allylic alcohol with DIBAL-H, asymmetric
Katsuki–Sharpless epoxidation using (2)-diisopropyl-
tartrate and deoxygenation by Cp2TiCl.11 Homoallylic
alcohol 4 was obtained from D-glyceraldehyde acetonide
as described by Roush.12 The two alcohols were coupled as
follows: 4 was silylated using excess dimethyldichloro-
silane and after evaporation of the volatile material the
Over the last 15 years, numerous research efforts have been
devoted to the study of intramolecular halogeno (seleno)
etherifications.3 In simple homoallylic alcohols, the 5-endo
Keywords: electrophilic; cyclization; oxetane; tetrahydrofuran.
*
Corresponding author. Tel.: þ33-3-89-33-68-58; fax: þ33-3-89-33-68-
60; e-mail: j.eustache@uha.fr
0040–4020/$ - see front matter q 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)01134-7

7366
P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
Figure 1. (Z)-2-ene-1,5-diols as versatile synthetic intermediates.
Figure 2. Possible cyclization modes in (Z)-2-ene-1,5-diols.
Figure 3. Electrophile-induced cyclization of homoallylic alcohols.
Scheme 1. Reagents and conditions. (a) Ph3PvCHCO2Me (1.5 equiv.), THF, 08C to rt, 16 h, 78%. (b) DIBAH (3 equiv.), toluene, 2788C, 2 h, 85%. (c)
Ti(OiPr)4, tBuOOH, (2)-DIPT, 16 h. (d) Cp2TiCl2 (5 equiv.), Zn (5 equiv.), ZnCl2 (2 equiv.), THF, 15 min, 67% (2 steps). (e) (i) BuLi (1.1 equiv.), THF,
2788C, 10 min then Me2SiCl2 (5 equiv.), 2788C to rt, 1 h; (ii) volatiles removal; (iii) 3, imidazole (3 equiv.), THF, rt, 16 h, 92%; (f) [Ru]-b (20 mol%), C6H6,
reflux, 48 h, 68%; (g) TBAF, THF, rt, 30 min, 80%.

P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
7367
crude chlorodimethylsilyl ether was allowed to react with 3
to afford acyclic silalketal 5. The RCM conditions were
examined next: using the ‘first generation’ Grubbs’s catalyst
[Ru]-b, and low concentrations of the reagents (0.01 M), 5
was smoothly converted to cyclic silalketal 6.
† The absence of signals in the region 3.5–3 ppm, which
rules out the presence of an epoxide.
† The high d values for protons H3 and H5, (4.45 and
4.75 ppm, respectively) consistent with an oxetane and
making a tetrahydrofuran structure unlikely.
† The two H4 protons couple only with the two low field
protons H3 and H5 and not with the proton adjacent to
Se, which rules out both the oxirane and tetrahydrofuran
possibilities.
Noteworthy is the absence of a NOE effect between H3 and
H5, indicative of a trans relationship between these protons.
This is confirmed by the strong NOE effects between H5 and
H4a and H3 and H4b. The strong coupling between H5 and
H4b is characteristic of a trans diaxial relationship while the
two weak couplings between H4 and H3 define an equatorial
position for the latter. Thus, although well supported by
many hints from the literature, the assumption we made at
the onset of this work that electrophile-induced cyclization
of diol 7 should preferably lead to tetrahydrofurans 1 proved
to be wrong.
Raising the concentration to 0.1 M led to the formation of
significant amounts of macrocycle 8, (the result of a cross
coupling metathesis followed by RCM).13 The same side-
reaction occurred when using the more active ‘second
generation’ Grubbs’s catalyst [Ru]-c, regardless of the
concentration of the reagents. Removal of the silicon tether
by fluoride ion provided the desired (Z)-2-ene-1,5-diol (7) in
50% overall yield (from alcohol 4).
Faced with this unexpected result, we felt that a more
detailed study of (Z)- and (E)-2-ene-1,5-diols, N-PSP-
induced cyclization, was warranted. The aim of this study
was twofold: beside gaining interesting information on the
almost unprecedented electrophile-induced cyclization of
2-ene-1,5-diols, we were hoping that (E)-2-ene-1,5-diols
might behave differently from the (Z)-isomers, perhaps
leading to preferential tetrahydrofurans formation. Then, the
approach could still be viable for accessing isomers of A.
These would also be interesting in connection to acetogenin
synthesis.15 For that purpose, systematic variation of the
structural elements thought to play an important role for
the outcome of the cyclization, i.e. the configuration of the
olefinic double bond and the relative stereochemistries at
the homoallylic and allylic carbon atoms was undertaken
(Fig. 5).16
With precursor 4 in hand, we proceeded to the key
electrophilic cyclization shown in Scheme 2. Our initial
attempts using a variety of halogen-based electrophiles
(NBS, NIS, I2-collidine) were unsuccessful. In contrast,
selenoetherification, using N-phenylselenenylphthalimide
(N-PSP) proceeded smoothly but afforded, instead of the
expected tetrahydrofuran(s), or oxirane(s) the two oxetanes
9 and 10 in 45 and 30% yields, respectively. 10 was
quantitatively converted back to 9 by treatment with 2,2-
dimethoxypropane (DMP). Finally, Bu3SnH removal of the
selenium in 9 gave oxetane 11 in 70% yield.
The 1-(R)-trans-ene 12 was easily obtained, albeit only in
modest yield, by photoisomerization of 717 (Scheme 3), and
the 1-(S)-cis-isomer 13 was prepared as shown in Scheme 4.
Thus, ynone 16 was obtained by acylation of bis-
trimethylsilylacetylene with tridecanoyl chloride.18
Enantioselective reduction according to Brown19 then
treatment with LiAlH4 afforded cleanly the (S)-allylic
alcohol 19 in .90% ee as determined from the 1H
NMR spectrum of the corresponding Mosher’s ester.
Figure 4. Structure determination of oxetane 9.
The structure of 9 (then 11) was established by proton and
13C NMR (Figure 4) and confirmed by synthesis (see
Ref. 14). Relevant features of the NMR data include:
Scheme 2. Reagents and conditions. (a) N-PSP (1.7 equiv.), CSA (0.1 equiv.), CH2Cl2, 2788C to rt, 16 h, 9 (45%) and 10 (30%); (b) DMP, CSA (0.2 equiv.),
rt, 16 h, quantitative; (c) Bu3SnH (3.0 equiv.), AIBN (cat.), toluene, reflux, 16 h, 70%.

7368
P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
Figure 5. N-PSP-induced cyclization: the four possible 5-(R)-2-ene-1,5-diols.
Scheme 3. Reagents and conditions. (a) hn, (PhS)2 (cat.), dioxane/hexane, rt. 2 h, 47% (two cycles).
Scheme 4. Reagents and conditions. (a) Me3Si–CuC–SiMe3 (1 equiv.), AlCl3 (1 equiv.), CH2Cl2, 08C, 2 h, 86%; (b) (S)-alpineborane, 08C to rt, neat, 16 h;
(c) TBAF, THF, rt, 30 min, 75% (two steps); (d) LiAlH4 (4.0 equiv.), THF, 08C to rt, 72 h, 74%; (e) (i) BuLi (1.1 equiv.), THF, 2788C, 10 min then Me2SiCl2
(4 equiv.), –788C to rt, 1 h; (ii) volatiles removal; (iii) 19, imidazole (3 equiv.), THF, rt, 16 h, 97%; (f) [Ru]-b (20 mol%), toluene, reflux, 48 h; (g) TBAF,
THF, rt, 30 min, 51% (two steps).
Coupling with homoallylic alcohol 4 through a silicon
bridge was performed as described earlier. The resulting
silalketal was cyclized to 21 using [Ru]-b. Removal of
the silicon tether by fluoride ion afforded 13 in 51%
overall yield.
aluminium hydride. 14 was obtained in 60% yield along
with 18% recovered unreacted 24.
With the three ene-diols 12, 13 and 14 in hand21 we
proceeded to the electrophilic cyclization (Scheme 6). The
conditions were the same as those used for 7 (treatment with
N-PSP, for 16 h) except that, after completion of the
reaction and removal of most phthalimide by filtration,
excess DMP was added to the reaction mixture in order to
convert the deprotected diol back to the corresponding
isopropylidene.
The last required (E,1S)-2-ene-1,5-diol was prepared as
shown in Scheme 5: propargylic alcohol 18 was converted
to the tert-butyldiphenylsilyl (TPS) ether 22 which was
deprotonated and allowed to react with the known epoxide
2520 in presence of BF3·OEt2. Removal of the TPS group
gave yne-diol 24 in excellent (85%) overall yield. Finally,
24 was reduced to ene-diol 14 by treatment with lithium
Some trends are clearly apparent:
Scheme 5. Reagents and conditions. (a) TPSCl, Et3N, 4-DMAP, CH2Cl2, reflux, 16 h, quantitative; (b) (i) BuLi, THF, 2788C, 30 min; (ii) BF3OEt2, THF,
2788C, 30 min; (iii) 25, THF, 2788C, 3 h, 88%; (c) TBAF, THF, 208C, 16 h, 97%; (d) LiAlH4 (4 equiv.), THF, 0–208C, 16 h, 14 (60%) and 24 (18%).

P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
7369
Scheme 6. Reagents and conditions. (a) (i) N-PSP (1.5 equiv.), CSA (0.1 equiv.), CH2Cl2, 2788C to rt, 16 h; (ii) DMP, CSA (0.2 equiv.), rt, 16 h,
9þ26þ27þ28 (53%) 9:26:27:28¼2:55:8:27 or 30þ31 (20%) ratio 30:31 ,1:9 or 30þ33 (40%) 30:33¼2:8; (b) Bu3SnH (3.0 equiv.), AIBN (cat.), toluene,
reflux, 10 h, 29 (85%), 32 (80%) or 34 (65%).
† The (Z)-ene-diols (7 and 13) lead to the exclusive
formation of (3R,5R)-trans-oxetanes.
† The (E)-ene-diols (12 and 14) lead to mixtures of
tetrahydrofurans/oxetanes in which tetrahydrofurans
predominate.
† Starting from the (E,1R,5R)-2-ene-1,5-diol 12, one
observes a strong preference for tetrahydrofuran over
oxetane formation but a poor selectivity 6-(R) vs. 6-(S)
tetrahydrofurans. In contrast, 14 leads to the exclusive
formation of a single tetrahydrofuran and a single
oxetane.
non mechanistic, study risky. A tentative explanation is
shown Figure 6: according to Knight’s model, the more
stable transition state is the one which places the large (L)
substituent in equatorial position. Inspection of the model
suggests an unfavourable electrostatic interaction between
the incoming OH and OH-7 in the cyclisation of 12 (but not
of 14).
When compared to previous related work, the main
difference in our case is the increased steric bulk at position
7 of the ene-diol system which, in the case of 7 and 13 is
probably responsible for the exclusive formation of
oxetanes. The easier formation of trans (relative to cis)
oxetanes has some precedent in the literature and can be
explained by a disfavoured transition state when substitu-
ents are on the same side of the forming ring.14
The trans relationship between H3 and H5 in all oxetane
derivatives was established from NMR data, as described
for 9 (no NOE between H3 and H5, strong NOE between H5
and one of the H4, coupling pattern within the H3, H4, H5
system similar to that observed in 9). The absolute
configurations of the selenium-bearing carbon atoms are
tentatively assigned and correspond to those expected in
such cyclizations. Upon treatment with Bu3SnH the mixture
30þ31 furnished 32. Regarding tetrahydrofuran formation
from the (E)-ene-diols, the results are similar to those
previously reported and the model proposed by Knight
et al.6e correctly predicts the observed configurations.
Obviously, the stereochemistry at C7 is responsible for the
difference between the strong stereoselectivity observed
when cyclizing 14, and 12 but the reason for this dramatic
effect is difficult to explain and the lack of precedent in the
literature makes any conclusion derived from the present,
3. Conclusion
While attempting to prepare a synthetic intermediate for
acetogenin synthesis, involving electrophilic selenocycliza-
tion of a (Z)-2-ene-1,5-diol as a key step, we observed the
exclusive formation of oxetanes, instead of the expected
tetrahydrofurans. This led us to extend our studies to three
other ene-diols, varying the configuration of the olefinic
double bond and the relative configuration of the two
hydroxyl groups (Z and E, syn and anti). From this work, the
following conclusions can be drawn: (E)-2-ene-1,5-diols
afford tetrahydrofurans as expected but, unlike previously

7370
P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
Figure 6. Tentative model for electrophilic cyclization of (E)- and (Z)-2-ene-1,5-diols.
reported examples of electrophile-induced homoallylic
alcohol cyclizations, the (Z)-2-ene-1,5-diols used in the
4. Experimental
study furnished oxetanes only. We believe that steric effect
considerations can explain this preferential oxetane for-
mation and suggest that electronic effects may influence the
differing stereochemical course of the cyclization of syn-
and anti-(E)-2-ene-1,5-diols. Obviously, these conclusions
should be taken with caution. Beside the major products
which were considered in the study, many impurities were
formed during the cyclization which could be neither
isolated nor characterized. Clearly, more stable substrates
(without functions other than those directly involved in the
cyclisation) should be used for ‘real’ mechanistic studies.
The present work, however, indicates that subtle factors,
which remain to be precisely defined, govern the oxetane/
tetrahydrofuran ratio in electrophile-induced homoallylic
alcohols cyclizations. Given the synthetic potential of the
selenocyclization reaction, further studies along this line are
warranted.
4.1. General
1H and 13C NMR spectra were recorded at 400 and 100 MHz,
respectively. Melting points were taken on a capillary melting
point apparatus and are not corrected. Optical rotations were
measured using a Perkin–Elmer 341 LC polarimeter. All
reactions were carried-out under argon atmosphere with
standard techniques for the exclusion of air and moisture. All
solvents were dried before use (CH2Cl2, toluene and benzene
were distilled over CaH2, THF was distilled from sodium
benzophenone ketyl). TLC was performed using fluorescent
60F254 Merck coated plates. Column chromatography was
performed on Macherey–Nagel silica gel 60 (70–230 mesh)
with the mixture of solvents indicated in each case.
4.1.1. (2R,3R)-(3-Dodecyl-oxiranyl)-methanol (2). To a
˚
suspension of 4 A molecular sieves (1.50 g) in CH2Cl2

P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
7371
(30 mL) kept at 2258C was added a solution of Ti(OiPr)4
(400 mL, 0.20 equiv., 1.34 mmol) in CH2Cl2, (3 mL), and
(S,S)-DIPT (360 mL, 1.68 mmol, 0.25 equiv.) in 3 mL
CH2Cl2. After 15 min, a solution of allylic alcohol 1
(1.52 g, 1.0 equiv., 6.72 mmol) in CH2Cl2 (5 mL) was
added dropwise and the reaction mixture was stirred at
2258C for 30 min. Then excess TBHP (5.0–6.0 M in
decanes, 8 mmol, 1.60 mL) was slowly added. After 14 h at
2258C, the temperature was raised to 2158C and the
reaction was quenched with a solution of tartaric acid
(30%). The aqueous layer was extracted with CH2Cl2 and
the combined organic extracts were dried over MgSO4.
After purification by chromatography over silica gel
(cyclohexane/ethyl acetate, 8:2), epoxide 2 (1.20 g, 74%)
was obtained: mp 68–708C; [a]2D0¼þ26.0 (c 1.0, CHCl3);
1H NMR (CDCl3, 250 MHz) d (ppm) 0.87 (3H, broad t,
J¼6.0 Hz, CH3), 1.18–1.55 (22H, m, 11£CH2), 1.82 (1H,
broad t, J¼5.9 Hz, OH), 2.95 (2H, m, CH2O), 3.60 (1H, m,
CHO oxirane), 3.90 (1H, m, CHO oxirane); 13C NMR
(CDCl3, 62.9 MHz) d (ppm) 14.1, 22.7, 25.2, 29.3–29.6,
31.5, 31.9, 56.0, 58.4, 61.7; Anal. calcd for C15H30O2: C,
74.38; H, 12.40. Found: C, 74.11; H, 12.41.
11£CH2), 1.33 (3H, s, CH3), 1.44 (3H, s, CH3), 2.10–2.35
(2H, m, CH2, H-4), 3.73 (1H, t, J¼7.6 Hz, CH, H-1), 3.85
(1H, m, CH, H-3), 3.97 (1H, t, J¼7.6 Hz, CH, H-1), 4.09
(1H, m, CH, H-2), 4.21 (1H, m, CH, H-9), 5.00–5.16 (4H,
m, 2£CH2, H-6 and H-7), 5.73–5.93 (2H, 2£CH, H-5 and
H-8); 13C NMR (CDCl3, 62.9 MHz) d (ppm) 21.6, 21.5,
14.1, 22.7, 25.2, 25.3, 26.4, 29.3–29.6, 31.9, 37.5, 37.8,
65.5, 72.8, 73.7, 78.3, 109.0, 113.7, 117.1, 134.9, 141.4.
4.1.4. Cyclic silalketal (6). A solution of 5 (1.10 g,
2.4 mmol) and Grubbs’s catalyst [Ru]-b (0.20 g,
0.25 mmol, 0.1 equiv.) in benzene (240 mL) was stirred at
reflux for 16 h. Then more catalyst (0.20 g, 0.25 mmol,
0.1 equiv.), was added and stirring was continued for 48 h.
The solvent was removed under reduced pressure and the
residue was chromatographed (cyclohexane/ethyl acetate,
97:3) to give 0.60 g of 5 (68%). [a]2D0¼229.3 (c 0.75,
1
CHCl3); H NMR (CDCl3, 400 MHz) d (ppm) 0.16 (6H, s,
CH3SiCH3), 0.90 (3H, broad t, J¼6.8 Hz, CH3), 1.26–1.60
(22H, m, 11£CH2), 1.36 (3H, s, CH3), 1.44 (3H, s, CH3),
2.22 and 2.68 (2H, m, CH2, H-4), 3.73 (1H, t, J¼7.8 Hz,
CH, H-1), 3.98 (2H, m, 2£CH, H-1 and H-3), 4.10 (1H, q,
J¼6.6 Hz, CH, H-2), 4.36 (1H, q, J¼6.0 Hz, CH, H-7), 5.59
(1H, m, CH, H-5), 5.76 (1H, dd, J¼11.2, 5.9 Hz, CH, H-6);
13C NMR (CDCl3, 100 MHz) d (ppm) 22.5, 0.0, 14.1, 22.7,
25.4, 25.8, 26.6, 29.3, 29.5, 29.6, 29.6, 29.7, 31.4, 31.9,
37.9, 65.9, 70.8, 74.5, 79.2, 109.4, 126.5, 137.0.
4.1.2. (R)-Pentadec-1-en-3-ol (3). To a red solution of
Cp2TiCl2 (6.22 g, 25.0 mmol, 5.0 equiv.) and ZnCl2 (1.36 g,
10.0 mmol, 2.0 equiv.) in THF (90 mL), was added 1.63 g
(5.0 equiv., 25.0 mmol), of powdered Zn. The reaction
mixture was stirred for 1 h at room temperature at which
time the colour had turned to green. Epoxide 2 (1.0 equiv.,
5.01 mmol) in 10 mL THF was added and stirring was
continued for 15 min. Aqueous HCl (1.0 M, 20 mL) was
added and the mixture was extracted with ether. The organic
layer was successively washed with water, 10% aqueous
NaHCO3, water, brine, dried over MgSO4 and concentrated
under reduced pressure. The residue was chromatographed
over silica gel (cyclohexane/ethyl acetate, 8:2), to afford 3
(1.05 g, 67%, two steps): mp 28–308C; [a]2D0¼26.2 (c 0.8,
4.1.5. (Z,1R,5R)-1-(4(S)-2,2-Dimethyl-[1,3]dioxolan-4-yl)-
heptadec-3-ene-1,5-diol (7). To a solution of 6 (30 mg,
0.08 mmol) in THF (0.2 mL) at room temperature was
added 60 mL of TBAF (1.0 M in THF, 0.60 mmol,
2.2 equiv.). After 30 min, the solvent was removed and
the residue chromatographed (cyclohexane/ethyl acetate,
7:3) to yield 7, (8 mg, 80%), a colorless oil. [a]2D0¼þ10.0 (c
1
0.85, CHCl3); H NMR (CDCl3, 400 MHz) d (ppm) 0.87
(3H, broad t, J¼6.8 Hz, CH3), 1.18–1.68 (22H, m,
11£CH2), 1.36 (3H, s, CH3), 1.45 (3H, s, CH3), 2.22–2.55
(4H, m, 2£OH and CH2, H-4), 3.60 (1H, q, J¼5.8 Hz, CH,
H-3), 3.80 (1H, broad t, J¼7.7 Hz, CH, H-1), 4.0 (1H, t,
J¼7.7 Hz, CH H-1), 4.10 (1H, q, J¼6.5 Hz, CH, H-2), 4.40
(1H, q, J¼6.9 Hz, CH, H-7), 5.47–5.65 (2H, m, 2£CH, H-5
and H-6); 13C NMR (CDCl3, 100 MHz) d (ppm) 14.1, 22.7,
25.2, 25.4, 26.5, 29.4, 29.61, 29.63, 29.66, 29.68, 31.9, 32.1,
37.0, 65.9 (C1), 67.1, 71.3, 77.6, 109.6, 126.4, 136.5; Anal.
calcd for C22H42O4: C, 71.35; H, 11.35. Found: C, 70.71; H,
11.38.
1
CHCl3); H NMR (CDCl3, 250 MHz) d (ppm) 0.87 (3H,
broad t, J¼6.0 Hz, CH3), 1.14–1.54 (22Hþ1H, m, 11£CH2
and OH), 4.07 (1H, q, J¼6.2 Hz, CHOH), 5.17 (2H, m,
CH2v), 5.86 (1H, m, CHv); 13C NMR (CDCl3, 62.9 MHz)
d (ppm) 14.1, 22.7, 25.3, 29.3–29.6, 31.9, 37.0, 73.3, 114.5,
141.3; Anal. calcd for C15H30O: C, 79.65; H, 13.27. Found:
C, 79.67; H, 13.31.
4.1.3. Silalketal (5). A cold (2788C) solution of homo-
allylic alcohol 4 (0.466 g, 2.70 mmol) in THF (7 mL) was
placed under argon atmosphere. BuLi (1.6 M in hexanes,
1.85 mL, 2.97 mmol) was added dropwise and the solution
was stirred for 15 min. Freshly distilled Me2SiCl2 (1.60 mL,
5.0 equiv., 13.55 mmol) was added and the reaction mixture
was allowed to warm to room temperature. The solvents and
Me2SiCl2 were removed under reduced pressure and a
solution of allylic alcohol 3 (660 mg, 2.90 mmol, 1.07 equiv.)
and imidazole (610 mg, 9 mmol, 3 equiv.) in THF (7 mL)
was added. The resulting suspension was stirred at room
temperature for 16 h, hydrolyzed and extracted with ethyl
acetate. The organic layer was washed with brine and dried
over MgSO4. After chromatography silica gel (cyclohexane/
ethyl acetate, 95:5), silalketal 5 (1.10 g, 92%) was obtained
4.1.6. Cyclization of (Z,1R,5R)-1-(4(S)-2,2-Dimethyl-
[1,3]dioxolan-4-yl)-heptadec-3-ene-1,5-diol 7: oxetanes
9 and 10. To a cold (2788C) solution of diol 7 (50 mg,
0.13 mmol) in CH2Cl2 (2 mL), under argon, was added CSA
(3 mg, 0.1 equiv.) and N-PSP (70 mg, 0.23 mmol,
1.7 equiv.). The temperature was slowly allowed to reach
208C and the reaction mixture was stirred for 16 h. After
solvent removal, the residue was chromatographed (cyclo-
hexane/ethyl acetate, 1:1) to give 9 (32 mg, 45%) and 10
(20 mg, 30%).
1
Oxetane 9. [a]2D0¼28.2 (c 1.4, CHCl3); H NMR (CDCl3,
1
as a colorless oil: [a]2D0¼þ4.2 (c 1.2, CHCl3); H NMR
250 MHz) d (ppm) 0.88 (3H, t, J¼6.6 Hz, CH3), 1.20–1.30
(20H, m, 10£CH2), 1.38 (3H, s, CH3), 1.44 (3H, s, CH3),
1.74 (2H, m, CH2, H-8), 2.05 (1H, dd, J¼13.0, 5.6 Hz, CH,
(CDCl3, 250 MHz) d (ppm) 0.12 (6H, s, CH3SiCH3), 0.87
(3H, broad t, J¼6.3 Hz, CH3), 1.20–1.52 (22H, m,

7372
P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
H-4), 2.45 (1H, ddd, J¼13.0, 10.1, 4.3 Hz, CH, H-4), 3.19
(1H, d, J¼2.2 Hz, OH), 3.25 (1H, dd, J¼2.6, 1.7 Hz, CH, H-
6), 3.90–4.10 (4H, m, CH2 and 2£CH, H-1, H-2 and H-7),
4.45 (1H, dd, J¼4.3, 2.8 Hz, CH, H-3), 4.75 (1H, ddd,
J¼10.1, 5.6, 1.6 Hz, CH, H-5), 7.20 and 7.50 (5H, m, SePh);
13C NMR (CDCl3, 62.9 MHz) d (ppm) 14.1, 19.4, 22.7,
25.8, 28.6, 29.38, 29.59, 29.66, 31.9, 35.4, 39.0, 60.7, 61.0,
70.6, 75.0, 75.4, 82.4, 97.4, 127.0, 129.2, 131.1, 133.1.
78.74, 109.48, 126.72, 136.80; Anal. calcd for C22H42O4: C,
71.35; H, 11.35. Found: C, 70.97; H, 11.67.
4.1.9. (Z,1R,5S)-1-(4(S)-2,2-Dimethyl-[1,3]dioxolan-4-
yl)-heptadec-3-ene-1,5-diol (13). Prepared as 7: from 21
(0.240 g), 0.190 g of 13 (91%) were obtained as a white
solid: mp 31–328C; [a]2D0¼224.5 (c 1.9, CHCl3); 1H NMR
(CDCl3, 400 MHz) d (ppm) 0.86 (3H, broad t, J¼6.8 Hz,
CH3), 1.24 (18H, broad s, 9£CH2), 1.37 (3H, s, CH3), 1.42
(3H, s, CH3), 1.40 and 1.83 (2H, m, CH2, H-8), 2.05 (1H, m,
CH2, H-4), 2.39 (1H, m, CH2, H-4), 3.05 (2H, broad s,
2£OH), 3.50 (1H, m, CH, H-3), 3.70 (1H, m, CH2, H-1),
4.02 (2H, m, CH2 and CH, H-1 and H-2), 4.35 (1H, broad q,
J¼7.0 Hz, CH, H-7), 5.57 (2H, m, 2£CHv, H-5 and H-6);
13C NMR (CDCl3, 100 MHz) d (ppm) 14.10, 20.80, 22.68,
25.26, 25.44, 26.65, 29.03, 29.34, 29.61, 29.64, 31.68,
31.91, 36.85, 53.83, 65.97, 66.52, 71.44, 78.80, 109.61,
127.43, 136.53; Anal. calcd for C22H42O4: C, 71.35; H,
11.35. Found: C, 71.31; H, 11.42.
Oxetane 10. 1H NMR (CDCl3, 250 MHz) d (ppm) 0.88 (3H,
t, J¼6.6 Hz, CH3), 1.20–1.30 (20H, m, 10£CH2), 1.75 (2H,
m, CH2, H-8), 2.10 (1H, dd, J¼12.9, 5.6 Hz, CH, H-4), 2.45
(1H, ddd, J¼12.9, 10.0, 4.3 Hz, CH, H-4), 2.45 (1H, broad s,
OH), 3.10 (1H, d, J¼2.7 Hz, OH), 3.30 (1H, m, CH, H-6),
3.43 (1H, d, J¼3.6 Hz, OH), 3.90–4.15 (4H, CH2 and
2£CH, H-1, H-2 and H-7), 4.60 (1H, m, CH, H-3), 4.70 (1H,
m, CH, H-5), 7.20 and 7.50 (5H, m, SePh); 13C NMR
(CDCl3, 62.9 MHz) d (ppm) 14.1, 22.7, 25.9, 29.37, 29.58,
29.66, 29.68, 31.9, 35.6, 41.1, 60.6, 61.8, 74.1, 75.2, 81.9,
82.3, 127.2, 128.2, 130.7, 133.4.
4.1.10. (E,1R,5S)-1-(4(S)-2,2-Dimethyl-[1,3]dioxolan-4-
yl)-heptadec-3-ene-1,5-diol (14). To a cold (08C) solution
of 24 (0.330 g, 0.90 mmol) in THF (5 mL) was added
LiAlH4 (0.140 g, 3.60 mmol, 4.0 equiv.). The reaction
mixture was stirred overnight at 208C. After usual work-
up, and extraction with ether and evaporation of the solvent,
the crude product was purified by chromatography (cyclo-
hexane/ethyl acetate, 6:4) to give 14 as a white solid
(0.200 g, 60%) and recovered 24 (60 mg, 18%) which was
4.1.7. Deselenylation of oxetane 9: oxetane 11. To a
solution of 20 mg (0.038 mmol) of 9 in toluene (1 mL),
under argon, was added a crystal of AIBN and 30 mL
(3 equiv., 0.114 mmol) of Bu3SnH. After 10 h at reflux, the
solvent was removed and the residue chromatographed
(cyclohexane/ethyl acetate, 1:1) to yield 11 (10 mg, 70%).
[a]2D0¼29.7 (c 0.35, CHCl3); 1H NMR (CDCl3, 250 MHz) d
(ppm) 0.88 (3H, t, J¼6.6 Hz, CH3), 1.15–1.60 (24H, m,
12£CH2 and CH, H-6 and H-4), 1.38 (3H, s, CH3), 1.44 (3H,
s, CH3), 2.17 (1H, dd, J¼13.1, 5.6 Hz, CH, H-4), 3.58 (1H,
d, J¼0.8 Hz, OH), 3.80–4.10 (4H, m, CH2 and 2£CH, H-1,
H-2 and H-7), 4.40 (1H, t; J¼3.6 Hz, CH, H-3), 4.55 (1H,
tdd, J¼10.1, 5.5, 2.7 Hz, CH, H-5); 13C NMR (CDCl3,
62.9 MHz) d (ppm) 14.1, 19.8, 22.7, 25.4, 28.3, 29.38,
29.66, 29.70, 31.9, 37.4, 41.0, 42.7, 60.9, 70.4, 71.7, 74.0,
79.5, 97.6; EIMS m/z (rel. int.) 355 (3), 201 (19), 197 (16),
157 (9), 143 (25), 125 (12), 114 (9), 99 (24), 81 (35), 68
(100), 59 (46), 43 (36); Anal. calcd for C22H42O4: C, 71.35;
H, 11.35. Found: C, 70.25; H, 11.28.
1
recycled: mp 41–428C; [a]2D0¼þ8.5 (c 1.75, CHCl3); H
NMR (CDCl3, 400 MHz) d (ppm) 0.86 (3H, broad t,
J¼6.8 Hz, CH3), 1.25 (20H, broad s, 10£CH2), 1.36 (3H, s,
CH3), 1.42 (3H, s, CH3), 1.52 (2H, m, CH2, H-8), 1.77 (1H,
broad s, OH), 2.19 (2H, broad t, J¼6.5 Hz, CH2, H-4), 2.40
(1H, d, J¼4.5 Hz, OH), 3.57 (1H, m, CH, H-3), 3.73 (1H, m,
CH, H-2), 4.05 (3H, m, CH2 and CH, H-1 and H-7), 5.57
(1H, dd, J¼15.5, 6.5 Hz, vCH, H-5), 5.67 (1H, m, vCH,
H-6); 13C NMR (CDCl3, 100 MHz) d (ppm) 14.11, 22.69,
25.34, 25.49, 26.42, 29.35, 29.58, 29.61, 29.66, 31.92,
36.58, 37.29, 66.04, 71.80, 72.71, 78.51, 109.46, 126.41,
136.56; Anal. calcd for C22H42O4: C, 71.35; H, 11.35.
Found: C, 71.11; H, 11.22.
4.1.8. (E,1R,5R)-1-(4(S)-2,2-Dimethyl-[1,3]dioxolan-4-
yl)-heptadec-3-ene-1,5-diol (12). Argon was bubbled for
30 min through a solution of 7 (0.195 g, 0.53 mmol) in
hexane (40 mL) and dioxane (13 mL) (molarity of solution
0.01 M). Diphenyl disulphide (35 mg, 0.16 mmol, 0.3 equiv.)
was added. The reaction mixture was irradiated with a light
source (mercury lamp, 450 W, medium pressure, quartz) for
1.5 h to provide a mixture of diols 7 and 12. The reaction
mixture was treated by addition of Et3N (4 mL), then the
solvents were removed under reduced pressure and the two
stereoisomers were separated by flash chromatography
(cyclohexane/ethyl acetate, 7:3) to give 12 (91 mg, 47%)
and 7 (50 mg, 25%) which was recycled. Data for 12: mp
4.1.11. 1-Trimethylsilanyl-pentadec-1-yn-3-one (16). To
a solution of tridecanoic acid (4.30 g, 20 mmol) in toluene
(30 mL) kept under argon at 208C, was slowly added oxalyl
chloride (2.10 mL, 1.2 equiv., 24 mmol). After 45 min, the
volatile material was removed under reduced pressure and
the residue was dissolved in CH2Cl2 (30 mL) and cooled to
08C under argon. Bis-trimethylsilylacetylene (4.55 mL,
20 mmol) then AlCl3 (2.66 g, 20 mmol) were added. The
reaction mixture was stirred for 2 h at 08C then poured on
ice/H2O and extracted with hexane. The organic layers were
washed successively with a saturated solution of Na2CO3
then brine. After chromatography (cyclohexane/ethyl acet-
ate, 97:3) 16 was obtained as a colorless oil (5.10 g, 86%).
1H NMR (CDCl3, 400 MHz) d (ppm) 0.24 (9H, s, SiMe3),
0.87 (3H, broad t, J¼6.8 Hz, CH3), 1.28 (18H, s, 9£CH2),
1.65 (2H, m, CH2, H-5), 2.54 (2H, t, J¼7.3 Hz, CH2, H-4);
13C NMR (CDCl3, 100 MHz) d (ppm) 20.74, 14.12, 22.70,
23.80, 23.97, 28.92, 29.33, 29.35, 29.44, 29.60, 29.64,
31.93, 45.33, 97.56, 102.10; Anal. calcd for C18H34OSi: C,
73.47; H, 11.56. Found: C, 73.36; H, 11.67.
1
47–488C; [a]2D0¼þ3.7 (c 0.8, CHCl3); H NMR (CDCl3,
400 MHz) d (ppm) 0.87 (3H, t, J¼6.7 Hz, CH3), 1.10–1.70
(22H, m, 11£CH2), 1.36 (3H, s, CH3), 1.43 (3H, s, CH3),
2.17 (1H, t, J¼6.4 Hz, CH2, H-4), 2.28 (1H, broad s, OH),
2.78 (1H, broad s, OH), 3.57 (1H, broad s, CH, H-3), 3.57,
4.00 (2H, m, CH2, H-1), 4.10 (2H, m, 2£CH, H-2 and H-7),
5.60 (2H, m, 2£CH, H-5 and H-6); 13C NMR (CDCl3,
100 MHz) d (ppm) 14.10, 22.69, 25.37, 25.46, 26.61, 29.35,
29.62, 29.08, 31.92, 36.52, 37.12, 66.06, 71.85, 72.87,

P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
7373
4.1.12. (S)-1-Trimethylsilanyl-pentadec-1-yn-3-ol (17).
To a cold (08C) solution of (S)-Alpine borane (0.5 M in
THF, 14 mL, 7 mmol, 2.0 equiv.) 16 (1.036 g, 3.52 mmol)
was added. The THF was removed under reduced pressure
at 08C and the reaction mixture was stirred overnight at
room temperature then cooled to 08C. Et2O (5 mL) was
added followed by acetaldehyde and diethanolamine (1 mL
each). The precipitate which formed was filtered-off and the
filtrate partitioned in water/HCl 1 M (9:1) and Et2O.
Chromatography (cyclohexane to cyclohexane/ethyl acet-
ate, 95:5) afforded 17 as a colorless oil (0.899 g, 86%)
which was used as such for the next step. [a]2D0¼þ1.03 (c
4.1.16. Cyclic silalketal (21). Prepared as 6: from 20
(0.460 g), 21 (0.240 g, 55%) was obtained. 1H NMR
(CDCl3, 400 MHz) d (ppm) 0.09 and 0.12 (6H, s,
Me2Si), 0.87 (3H, broad t, J¼6.8 Hz, CH3), 1.40 (20H,
broad s, 10£CH2), 1.45 and 1.55 (2H, CH2, H-8), 2.34
(2H, m, CH2, H-4), 3.72 (1H, m, CH, H-3), 3.86 (1H, dd,
J¼8.30, 6.8 Hz, CH2, H-1), 4.02 (1H, dd, J¼8.3, 6.5 Hz,
CH2, H-1), 4.15 (1H, m, CH, H-2), 4.45 (1H, m, CH,
H-7), 5.68 (2H, m, 2£CHv, H-5 and H-6); 13C NMR
(CDCl3, 100 MHz) d (ppm) 21.31, 21.21, 14.11, 22.70,
25.35, 25.54, 26.37, 29.37, 29.55, 29.61, 29.64, 29.67,
30.67, 31.94, 37.32, 65.50, 67.91, 73.36, 78.64, 109.41,
128.93, 137.38.
1
1.85, CHCl3); H NMR (CDCl3, 400 MHz) d (ppm) 0.10
(9H, s, SiMe3), 0.81 (3H, broad t, J¼6.8 Hz, CH3), 1.15
(18H, s, 9£CH2), 1.30 (2H, m, CH2, H-5), 1.60 (2H, m, CH2,
H-4), 1.78 (1H, d, J¼5.6 Hz, OH), 4.28 (1H, broad q,
J¼5.7 Hz, CH, H-3); 13C NMR (CDCl3, 100 MHz) d (ppm)
0.00, 14.12, 22.71, 25.14, 29.25, 29.38, 29.53, 29.57, 29.67,
29.71, 32.00, 37.76, 62.95, 89.29, 107.04. HMRS m/z found
319.2415, calcd for C18H36ONaSi m/z 319.2433.
4.1.17. (S)-Pentadec-1-yn-3-ol, tert-butyldimethylsilyl
ether (22). To a solution of 18 (0.700 g, 3.125 mmol) in
CH2Cl2 (6 mL) under argon were successively added
Et3N (525 mL, 3.75 mmol, 1.2 equiv.), DMAP (0.190 g,
1.56 mmol, 0.5 equiv.) then TPSCl (895 mL, 3.44 mmol,
1.1 equiv.). The reaction mixture was refluxed for 16 h,
quenched with water and extracted with CH2Cl2. After
chromatography (cyclohexane/ethyl acetate, 95:5) 22 was
isolated as a white solid (1.44 g, quantitative): mp 34–
358C; [a]2D0¼221.2 (c 0.8, CHCl3); 1H NMR (CDCl3,
400 MHz) d (ppm) 0.80 (3H, broad t, J¼6.8 Hz, CH3),
1.05 (9H, s, tBuSi), 1.10 (18H, s, 9£CH2), 1.25 (2H, m,
CH2, H-5), 1.60 (2H, m, CH2, H-4), 2.24 (1H, broad s,
CH, H-1), 4.27 (1H, m, CH, H-3), 7.37 and 7.70 (10H, m,
2£PhSi); 13C NMR (CDCl3, 100 MHz) d (ppm) 14.16,
19.32, 22.73, 24.69, 26.59, 26.90, 26.95, 29.21, 29.40,
29.50, 29.57, 29.67, 29.69, 29.72, 31.97, 38.27, 63.78,
72.51, 85.20, 127.42, 127.60, 129.52, 129.65, 133.60,
133.73, 135.57, 135.88.
4.1.13. (S)-Pentadec-1-yn-3-ol (18). To a solution of 17 in
4 mL THF was added TBAF (1.0 M in THF, 4 mL,
4 mmol). The solution was stirred 30 min at rt. The solvent
was removed under reduced pressure and the residue was
purified by chromatography (cyclohexane/ethyl acetate,
7:3) to yield 18 as a white solid (0.590 g, 75%): mp 34–
358C; [a]2D0¼21.9 (c 1.15, CHCl3); 1H NMR (CDCl3,
400 MHz) d (ppm) 0.87 (3H, broad t, J¼6.6 Hz, CH3), 1.20
(18H, s, 9£CH2), 1.42 (2H, m, CH2, H-5), 1.70 (2H, m, CH2,
H-4), 2.13 (1H, d, J¼5.2 Hz, OH), 2.44 (1H, d, J¼2 Hz, CH,
H-1), 4.36 (1H, m, CH, H-3); 13C NMR (CDCl3, 100 MHz)
d (ppm) 14.10, 22.68, 25.03, 29.25, 29.35, 29.52,
29.54, 29.64, 29.67, 31.92, 37.65, 62.31, 72.77, 85.09;
Anal. calcd for C15H28O: C, 80.36; H, 12.50. Found: C,
80.11; H, 12.56.
4.1.18. (1R,5S)-1-(4(S)-2,2-Dimethyl-[1,3]dioxolan-4-yl)-
5-trimethylsilanyloxy-heptadec-3-yn-1-ol (23). To a cold
(2788C) solution of silyl ether 22 (0.770 g, 1.66 mmol,
1.2 equiv.) in THF (10 mL) was added dropwise BuLi
(1.6 M in hexanes, 1.15 mL, 1.81 mmol, 1.3 equiv.).
Stirring was continued for 30 min and BF3·OEt2 (230 mL,
1.81 mmol, 1.3 equiv.) was added. After a further 30 min 25
(0.200 g, 1.39 mmol) dissolved in THF (5 mL) was
transferred to the solution and stirring was continued for
2 h at which time the reaction was stopped by addition of a
saturated solution of NH4Cl. The mixture was allowed to
warm to 208C and was extracted by ethyl acetate. After
usual work-up and solvent removal under reduced pressure.
The crude product was purified by chromatography
(cyclohexane/ethyl acetate, 8:2) to give 23 as a colorless
4.1.14. (S)-Pentadec-1-en-3-ol (19). The propargylic alco-
hol 18 (0.400 g, 1.77 mmol) was added to a suspension of
LiAlH4 (0.270 g, 7.10 mmol, 4.0 equiv.) in cold (08C) THF.
The reaction mixture was stirred for 3 days at 208C, then
slowly hydrolyzed. After extraction with ethyl acetate, the
combined organic layers were dried over MgSO4 and the
solvents evaporated. The crude product was chromato-
graphed (cyclohexane/ethyl acetate, 95:5) to give 19 as a
white solid (300 mg, 74%): mp 28–308C; [a]2D0¼þ6.0
(c 0.7, CHCl3); spectroscopic data as for 3.
4.1.15. Silalketal (20). Prepared as 5: from 0.200 g of
homoallylic alcohol 4 and 0.300 g of allylic alcohol 19,
0.510 g (97%) of silaketal 20 were isolated. [a]2D0¼þ11.9
(c 1.6, CHCl3); 1H NMR (CDCl3, 400 MHz) d (ppm) 0.13
(6H, broad s, Me2Si), 0.87 (3H, broad t, J¼6.6 Hz, CH3),
1.25 (20H, broad s, 10£CH2), 1.34 (3H, s, CH3), 1.41
(3H, s, CH3), 1.50 (2H, m, CH2, H-10), 2.30 (2H, m, CH2,
H-4), 3.73 (1H, broad t, J¼7.7 Hz, CH2, H-1), 3.85 (1H,
m, CH, H-3), 3.95 (1H, m, CH2, H-1), 4.08 (1H, m, CH,
H-2), 4.20 (1H, m, CH, H-9), 5.10 (4H, m, 2£CH2v, H-6
and H-7), 5.80 (2H, m, 2£CHv, H-5 and H-8); 13C NMR
(CDCl3, 100 MHz) d (ppm) 21.68, 21.34, 14.12 22.71,
25.27, 25.37, 26.45, 29.38, 29.67, 31.94, 37.60, 37.87,
65.56, 72.86, 73.72, 78.42, 109.14, 113.82, 117.15,
134.90, 141.36.
1
oil (0.740 g, 88%). [a]2D0¼237.3 (c 1.5, CHCl3); H NMR
(CDCl3, 400 MHz) d (ppm) 0.90 (3H, broad t, J¼6.8 Hz,
CH3), 1.08 (9H, s, tBuSi), 1.15–1.40 (20H, m, 10£CH2),
1.35 (3H, s, CH3), 1.40 (3H, s, CH3), 1.67 (2H, m, CH2, H-
8), 2.16 (1H, d, J¼6 Hz, OH), 2.27 (2H, m, CH2, H-4), 3.43
(1H, m, CH, H-3), 3.74 and 3.92 (2H, m, CH2, H-1), 4.02
(1H, m, CH, H-2), 4.37 (1H, m, CH, H-7), 7.40 and 7.70
(10H, m, 2£PhSi); 13C NMR (CDCl3, 100 MHz) d (ppm)
14.12, 19.29, 22.70, 24.12, 24.99, 25.22, 26.48, 26.89,
29.24, 29.37, 29.55, 29.59, 29.66, 29.70, 31.94, 38.54,
64.03, 65.97, 70.31, 77.43, 80.47, 84.17, 109.35, 127.32,
127.55, 129.53, 129.68, 133.80, 134.13, 135.81, 135.99.
HMRS m/z found 629.4019, calcd for C38H58O4NaSi m/z
629.4002.

7374
P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
4.1.19. (1R,5S)-1-(4(S)-2,2-Dimethyl-[1,3]dioxolan-4-yl)-
heptadec-3-yne-1,5-diol (24). A solution of 23 (0.580 g,
0.96 mmol) in THF (3 mL) was treated by TBAF (1.0 M in
THF, 1.7 mL, 1.7 mmol, 1.5 equiv.) for 16 h at rt. The
solvent was removed under reduced pressure and the residue
chromatographed (cyclohexane/ethyl acetate, 1:1) to give
24 as a white solid (0.350 g, 97%): mp 34–358C;
31.95, 32.39, 37.36, 39.65, 66.29, 73.41, 77.13, 77.50,
88.77, 109.74, 127.97, 129.25, 134.70. HMRS m/z found
549.2518, calcd for C28H46O4NaSe m/z 549.2459.
4.1.21. Deselenylation of tetrahydrofuran 26: tetrahy-
drofuran 29. The same procedure as for the deselenylation
of 9 was used. From 140 mg (0.27 mmol) of 26 were
isolated 83 mg of 29 (85% yield, cyclohexane/ethyl acetate,
1:1). [a]2D0¼þ13.3 (c 0.75, CHCl3).22 1H NMR (CDCl3,
400 MHz) d (ppm) 0.87 (3H, t, J¼6.8 Hz, CH3), 1.15–1.60
(22H, m, 11£CH2), 1.38 (3H, s, CH3), 1.43 (3H, s, CH3),
1.65 and 2.05 (4H, m, 2£CH2, H-4 and H-5), 2.38 (1H, s,
OH), 3.40 (1H, m, CH, H-7), 3.70 (1H, t, J¼7.8 Hz, CH2, H-
1), 3.85 (1H, m, CH, H-6), 4.00 (2H, m, CH2 and CH, H-1
and H-3), 4.10 (1H, m, CH, H-2); 13C NMR (CDCl3,
100 MHz) d (ppm) 14.13, 22.71, 25.60, 25.67, 26.49, 28.25,
28.49, 29.37, 29.62, 29.64, 29.67, 29.67, 29.69, 29.74,
31.94, 33.49, 65.96, 74.05, 78.62, 79.62, 83.40, 109.80;
EIMS m/z (rel. int.) 355 (6), 269 (38), 171 (31), 113 (100),
101 (49), 83 (51), 69 (38), 57 (57), 43 (87).
1
[a]2D0¼22.4 (c 0.9, CHCl3); H NMR (CDCl3, 400 MHz):
d (ppm): 0.87 (3H, broad t, 6.8 Hz, CH3), 1.25 (18H, broad
s, 9£CH2), 1.37 (3H, s, CH3), 1.39 (2H, m, CH2, H-9),1.43
(3H, s, CH3), 1.67 (2H, m, CH2, H8), 2.29 (1H, broad s,
OH), 2.48 (2H, m, CH2, H-4), 2.67 (1H, d, J¼5.5 Hz, OH),
3.70 (1H, m, CH, H-3), 3.83 and 4.05 (2H, m, CH2, H-1),
4.15 (1H, m, CH, H-2), 4.34 (1H, m, CH, H-7); 13C NMR
(CDCl3, 100 MHz): d (ppm): 14.12, 22.70, 24.11, 25.21,
25.25, 26.52, 29.31, 29.36, 29.58, 29.60, 29.66, 29.68,
31.93, 38.00, 62.57, 66.02, 70.51, 77.59, 80.57, 83.93,
109.57; Anal. calcd for C22H40O4: C, 71.74; H, 10.87.
Found: C, 71.76; H, 11.03.
4.1.20. Cyclization of (E,1R,5R)-1-(4(S)-2,2-Dimethyl-
[1,3]dioxolan-4-yl)-heptadec-3-ene-1,5-diol 12: oxetanes
9 and 27, tetrahydrofurans 26 and 28. The same pro-
cedure as for the cyclization of 7 was used. From 180 mg
(0.486 mmol) of 12 were obtained 9þ27 (21 mg), 28
(38 mg) and (26) 75 mg (combined yield: 53%—cyclohexane/
ethyl acetate, 1:1).
4.1.22. Cyclization of (Z,1R,5S)-1-(4(S)-2,2-Dimethyl-
[1,3]dioxolan-4-yl)-heptadec-3-ene-1,5-diol 13: oxetanes
30 and 31. The same procedure as for the cyclization of 7
was used. From 13 (40 mg) were obtained: 10 mg (20%) of
a 1:9 mixture of 30 and 31 from which pure 31 could be
isolated by chromatography (cyclohexane/ethyl acetate,
1:1), [a]2D0¼28.9 (c 0.95, CHCl3); 1H NMR (CDCl3,
400 MHz) d (ppm) 0.87 (3H, t, J¼6.7 Hz, CH3), 1.20
(18H, s, 9£CH2), 1.36 (3H, s, CH3), 1.42 (3H, s, CH3), 1.40
and 1.70 (2H, m, CH2, H-8), 2.10 (1H, dd, J¼13.1, 5.5 Hz,
CH2, H-4), 2.38 (1H, ddd, J¼13.1, 10.3, 4.3 Hz, CH2, H-4),
2.83 (1H, broad d, J¼6.8 Hz, OH), 3.30 (1H, d, J¼3.0 Hz,
CH, H-6), 3.95 (4H, m, CH2 and 2£CH, H-1, H-2 and H-7),
4.46 (1H, m, CH, H-3), 4.82 (1H, ddd, J¼10.3, 5.6, 1.5 Hz,
CH, H-5), 7.25 and 7.60 (5H, m, SePh); 13C NMR (CDCl3,
100 MHz) d (ppm) 14.13, 19.70, 22.70, 25.47, 26.14, 28.38,
29.37, 29.53, 29.62, 29.66, 29.69, 31.94, 35.67, 39.19,
58.92, 60.95, 70.64, 74.69, 75.44, 78.03, 97.53, 127.27,
129.19, 130.56, 133.52.
1
Tetrahydrofuran 26. H NMR (CDCl3, 400 MHz) d (ppm)
0.83 (3H, t, J¼6.8 Hz, CH3), 1.25–1.60 (22H, m, 11£CH2),
1.33 (3H, s, CH3), 1.37 (3H, s, CH3), 1.80 (1H, broad s,
OH), 1.90 (1H, m, CH2, H-4), 2.37 (1H, m, CH2, H-4),
3.50 (1H, m, CH, H-7), 3.60 (2H, m, CH2 and CH, H-1
and H-5), 3.75 (1H, m, CH, H-6), 3.90 (2H, m, CH2 and
CH, H-1 and H-3), 4.05 (1H, m, CH, H-2), 7.20 and 7.50
(5H, m, PhSe); 13C NMR (CDCl3, 100 MHz) d (ppm)
14.13, 22.70, 25.58, 25.89, 26.38, 29.37, 29.50, 29.62,
29.67, 29.70, 31.95, 35.03, 36.68, 39.79, 65.73, 70.44,
78.17, 79.31, 86.74, 109.84, 128.11, 129.21, 135.16;
Anal. calcd for C28H46O4Se: C, 64.00; H, 8.76. Found: C,
64.09; H, 8.85.
4.1.23. Deselenylation of the mixture of oxetanes 30 and
31: oxetane 32. The same procedure as for the deselenyla-
tion of 9 was used. From 15 mg of 31 were isolated 8 mg of
32 (80% yield, cyclohexane/ethyl acetate, 1:1). White solid:
mp 53–548C; [a]D20¼27.5 (c¼0.4, CHCl3); 1H NMR
(CDCl3, 400 MHz) d (ppm) 0.86 (3H, t, J¼6.8 Hz, CH3),
1.20 (20H, s, 10£CH2), 1.33 (3H, s, CH3),1.40 (2H, m, CH2,
H-9),1.41 (3H, s, CH3), 1.60 and 1.80 (5H, m, 3£CH2, H-4,
H-6 and H-8), 2.10 (1H, dd, J¼13.1, 5.3 Hz, CH2, H-4),
2.69 (1H, broad s, OH), 3.80 (3H, m, CH2 and 2£CH, H-1,
H-2 and H-7), 4.05 (1H, dd, J¼12.3, 3.3 Hz, CH2, H-1),
4.45 (1H, broad t, J¼3.5 Hz, CH, H-3), 4.60 (1H, m, CH, H-
5); 13C NMR (CDCl3, 100 MHz) d (ppm) 14.13, 20.13,
22.70, 25.80, 28.03, 29.37, 29.63, 29.65, 29.69, 31.94,
37.48, 39.98, 41.40, 60.87, 69.34, 70.95, 74.09, 76.58,
97.70. HMRS m/z found 393.2956, calcd for C22H42O4Na
m/z 393.2981.
Oxetane 27. 1H NMR (CDCl3, 400 MHz) d (ppm) 0.87 (3H,
t, J¼6.8 Hz, CH3), 1.05–1.60 (22H, m, 11£CH2), 1.42 (3H,
s, CH3), 1.46 (3H, s, CH3), 1.70 (1H, m, CH2, H-4), 2.45
(1H, m, CH2, H-4), 3.10 (1H, d, J¼5.5 Hz, OH), 3.25 (1H,
m, CH, H-6), 4.00 (4H, m, CH2 and 2£CH, H-1, H-2 and
H-7), 4.35 (1H, m, CH, H-3), 4.72 (1H, m, CH, H-5), 7.28
and 7.55 (5H, m, SePh); 13C NMR (CDCl3, 100 MHz) d
(ppm) 14.13, 19.62, 22.70, 25.72, 28.43, 29.37, 29.58,
29.65, 29.67, 29.70, 31.94, 35.04, 40.95, 59.54, 60.91,
70.42, 72.99, 74.41, 80.75, 97.56, 127.68, 129.34, 134.48,
134.79.
1
Tetrahydrofuran 28. H NMR (CDCl3, 400 MHz) d (ppm)
0.87 (3H, t, J¼6.8 Hz, CH3), 1.15–1.50 (22H, m, 11£CH2),
1.37 (3H, s, CH3), 1.44 (3H, s, CH3), 1.80 (1H, broad s, OH),
2.00 (1H, m, CH2, H-4), 2.35 (1H, m, CH2, H-4), 3.75 (1H,
m, CH, H-7), 3.84 (1H, m, CH2, H-1), 3.93 (1H, m, CH, H-
5), 4.00 (2H, m, CH2 and CH, H-1 and H-6), 4.12 (1H, m,
CH, H-2), 4.20 (1H, m, CH, H-3), 7.20 and 7.50 (5H, m,
PhSe); 13C NMR (CDCl3, 100 MHz) d (ppm) 14.14, 22.72,
25.73, 26.05, 29.39, 29.50, 29.62, 29.68, 29.71, 30.95,
4.1.24. Cyclization of (E,1R,5S)-1-(4(S)-2,2-Dimethyl-
[1,3]dioxolan-4-yl)-heptadec-3-ene-1,5-diol 14: oxetanes
30 and tetrahydrofuran 33. The same procedure as for the
cyclization of 7 was used. From 43 mg (0.48 mmol) of 14

P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
7375
references therein. For a recent review see: (c) Robin, S.;
Rousseau, G. Eur. J. Org. Chem. 2002, 3099–3114.
were obtained oxetane 30 (5 mg) and tetrahydrofuran 33
(20 mg) (combined yield: 41%—cyclohexane/ethyl acetate,
1:1).
4. See for example: Evans, R. D.; Magee, J. W.; Schauble, J. H.
Synthesis 1988, 862–868.
Oxetane 30. 1H NMR (CDCl3, 400 MHz) d (ppm) 0.88 (3H,
t, J¼6.8 Hz, CH3), 1.25 (20H, s, 10£CH2), 1.36 (3H, s,
CH3), 1.42 (3H, s, CH3), 1.45 and 1.68 (2H, m, CH2, H-8),
1.76 (1H, ddd, J¼13.6, 9.8, 3.0 Hz, CH2, H-4), 2.32 (1H, dd,
J¼13.6, 5.6 Hz, CH2, H-4), 2.70 (1H, broad s, OH), 3.30
(1H, dd, J¼8.3, 2.5 Hz, CH, H-6), 3.86–4.00 (4H, m, CH2
and 2£CH, H-1, H-2 and H-7), 4.37 (1H, broad t, J¼3.2 Hz,
CH, H-3), 4.72 (1H, ddd, J¼9.85, 8.6, 5.6 Hz, CH, H-5),
7.25 and 7.60 (5H, m, SePh); 13C NMR (CDCl3, 100 MHz)
d (ppm) 14.13, 19.73, 22.71, 25.63, 26.91, 29.38, 29.54,
29.63, 29.67, 29.70, 31.95, 34.71, 40.35, 58.65, 60.90,
67.99, 70.58, 72.85, 74.45, 80.69, 97.62, 127.49, 129.16,
129.99, 133.85, 134.26.
5. Baldwin, J. E.; Cutting, J.; Dupont, W.; Kruse, L.; Silberman,
L.; Thomas, R. C. J. Chem. Soc., Chem. Commun. 1976,
736–738.
6. (a) Bedford, S. B.; Bell, K. E.; Fenton, G.; Hayes, C. J.;
Knight, D. W.; Shaw, D. E. Tetrahedron Lett. 1992, 33,
6511–6514. (b) Galatsis, P.; Millan, S. D.; Nechala, P.;
Ferguson, G. J. Org. Chem. 1994, 59, 6643–6651. (c) Jung,
M. E.; Nichols, C. J. Tetrahedron Lett. 1996, 37, 7667–7670.
(d) Galatsis, P.; Millan, S. D.; Ferguson, G. J. Org. Chem.
1997, 62, 5048–5056. (e) Fujita, K.-I.; Murata, K.; Iwaoka,
M.; Tomoda, S. Tetrahedron 1997, 53, 2029–2048. (f)
Andrey, O.; Ducry, L.; Landais, Y.; Planchenault, D.;
Weber, V. Tetrahedron 1997, 53, 4339–4352. (g) Bedford,
S. B.; Bell, K. E.; Bennet, F.; Hayes, C. J.; Knight, D. W.;
Shaw, D. E. J. Chem. Soc., Perkin Trans 1 1999, 2143–2153.
7. (a) Barks, J. M.; Knight, D. W.; Seaman, C. J.; Weingarten,
G. G. Tetrahedron Lett. 1994, 35, 7259–7262. (b) In contrast,
exclusive oxetane formation is observed for homoallylic
alcohols in which a silicon atom is attached to the olefinic
carbon atom distal from the hydroxyl group: Rofoo, M.; Roux,
M.-C.; Rousseau, G. Tetrahedron Lett. 2001, 42, 2481–2484.
8. Temesvari-Major, E.; Gruber, L.; Toemoeskoezi, I.; Kovacs,
G.; Cseh, Gy. Tetrahedron Lett. 1980, 21, 4035–4038.
9. For previous examples of tethered ring-closing metathesis
using silalketals see: (a) Evans, P. A.; Murthy, V. S. J. Org.
Chem. 1998, 63, 6768–6769. (b) Hoye, T. R.; Promo, M. A.
Tetrahedron Lett. 1999, 40, 1429–1432. (c) Briot, A.; Bujard,
M.; Gouverneur, V.; Nolan, S. P.; Mioskowski, C. Org. Lett.
2000, 2, 1517–1519. (d) Gierash, T. M.; Chytil, M.; Didiuk,
M. T.; Park, J. Y.; Urban, J. J.; Nolan, S. P.; Verdine, G. L.
1
Tetrahydrofuran 33. H NMR (CDCl3, 400 MHz) d (ppm)
0.87 (3H, broad t, J¼6.8 Hz, CH3), 1.25 (20H, broad s,
10£CH2), 1.36 (3H, s, CH3), 1.42 (3H, s, CH3), 1.50 (2H, m,
CH2, H-8), 1.83 (1H, m, CH2, H-4), 2.05 (1H, broad s, OH),
2.42 (1H, m, CH2, H-4), 3.55 (1H, dd, J¼7.9, 7.0 Hz, CH,
H-2), 3.74 (1H, m, CH, H-7), 3.95–4.10 (4H, m, CH2 and
2£CH, H-1, H-3 and H-6), 7.28 and 7.55 (5H, m, PhSe); 13C
NMR (CDCl3, 100 MHz) d (ppm) 14.13, 22.71, 25.44,
26.20, 26.57, 29.38, 29.54, 29.56, 29.63, 29.67, 29.70,
31.94, 32.53, 37.45, 38.01, 65.79, 78.12, 80.33, 86.87,
110.01, 123.63, 128.10, 128.52, 129.23, 134.34, 135.05.
HMRS m/z found 549.2470, calcd for C28H46O4NaSe m/z
549.2459.
4.1.25. Deselenylation of tetrahydrofuran 33: tetrahy-
drofuran 34. The same procedure as for the deselenylation
of 9 was used. From 25 mg of 33 were isolated 11 mg of 34
(65% yield, cyclohexane/ethyl acetate, 1:1). [a]2D0¼þ8.0
¨
Org. Lett. 2000, 2, 3999–4000. (e) Lobbel, M.; Koll, P.
Tetrahedron: Asymmetry 2000, 11, 393–396. (f) Harrison,
B. A.; Verdine, G. L. Org. Lett. 2001, 3, 2157–2159. (g)
Harrison, B. A.; Gierash, T. M.; Neilan, C.; Pasternak, G. W.;
Verdine, G. L. J. Am. Chem. Soc 2002, 124, 13352–13353.
10. For previous examples of tethered ring-closing metathesis
using silylethers see: (a) Chang, S.; Grubbs, R. H. Tetrahedron
Lett. 1997, 38, 4757–4760. (b) Meyer, C.; Cossy, J.
Tetrahedron Lett. 1997, 38, 7861–7864. (c) Cassidy, J. H.;
Marsden, S. P.; Stemp, G. Synlett 1997, 1411–1413. (d)
Ahmed, M.; Barrett, A. G. M.; Beall, J. C.; Braddock, D. C.;
Flack, K.; Gibson, V. C.; Procopiou, P. A.; Salter, M. M.
Tetrahedron 1999, 55, 3219–3232. (e) Hoshi, T.; Yasuda, H.;
Sanji, T.; Sakurai, H. Bull. Chem. Soc. Jpn 1999, 72, 821–827.
(f) Taylor, R. E.; Engelhardt, F. C.; Yuan, H. Org. Lett. 1999,
1, 1257–1260. (g) Denmark, S. E.; Yang, S.-M. Org. Lett.
2001, 3, 1749–1752. (h) Yao, Q. Org. Lett. 2001, 3,
2069–2072. (i) Denmark, S. E.; Yang, S.-M. J. Am. Chem.
Soc 2002, 124, 15196–15197.
1
(c 0.5, CHCl3); H NMR (CDCl3, 400 MHz) d (ppm) 0.87
(3H, broad t, J¼6.8 Hz, CH3), 1.20 (20H, broad s, 10£CH2),
1.32 (3H, s, CH3), 1.43 (3H, s, CH3), 1.40–1.55 (2H, m,
CH2, H-8), 1.55–1.95 (4H, m, 2£CH2, H-4 and H-5), 2.10
(1H, broad s, OH), 3.62 (1H, t, J¼7.3 Hz, CH2, H-1), 3.85–
4.05 (5H, m, CH2, 4£CH, H-1, H-2, H-3, H-6 and H-7); 13
C
NMR (CDCl3, 100 MHz) d (ppm) 14.13, 22.70, 24.42,
25.56, 26.04, 26.57, 28.38, 29.37, 29.56, 29.62, 29.66,
29.68, 31.94, 32.48, 65.91, 71.17, 79.00, 80.51, 82.86,
109.87. HMRS m/z found 393.2976, calcd for C22H42O4Na
m/z 393.2981.
References
1. (a) Boiteau, J.-G.; Van de Weghe, P.; Eustache, J. Tetrahedron
Lett. 2001, 42, 239–242. (b) Van de Weghe, P.; Aoun, D.;
Boiteau, J.-G.; Eustache, J. Org. Lett. 2002, 4, 4105–4108.
2 A would give rise to rare 2(R), 5(S) acetogenins (see Figure 1
for numbering). The corresponding 2(R), 5(R) isomer has been
used as synthetic intermediate in the synthesis of corossolins:
Yu, Q.; Yao, Z. J.; Chen, X. G.; Wu, Y. L. J. Org. Chem. 1999,
64, 2440–2445.
11. Yadav, J. S.; Valli, M. H.; Prasad, A. R. Tetrahedron 1998, 54,
7551–7562.
12. Roush, W. R.; Hoong, L. K.; Palmer, M. A. J.; Park, J. C.
J. Org. Chem. 1990, 55, 4109–4117.
13. This behaviour is not surprising and has already been observed
for difficult RCMs. See for example: (a) Joe, D.; Overman,
L. E. Tetrahedron Lett. 1997, 38, 8635–8638. (b) Nevalainen,
M.; Koskinen, A. M. P. Angew. Chem. Int. Ed. 2001, 40,
4060–4062.
´
3. (a) Bravo, F.; Castillon, S. Eur. J. Org. Chem. 2001, 507–516,
and references therein. (b) Bew, S. P.; Barks, J.; Knight, D. W.;
Middleton, R. J. Tetrahedron Lett. 2000, 41, 4447–4451, and
14. In cis-2,4-disubstituted oxetanes, there is a strong NOE effect

7376
P. Van de Weghe et al. / Tetrahedron 59 (2003) 7365–7376
between H2 and H4 see: Aftab, T.; Carter, C.; Chrstlieb, M.;
acetogenins, see: Maezaki, N.; Kojima, N.; Asai, M.;
Tominaga, H.; Tanaka, T. Org. Lett. 2002, 4, 2977–2980.
17. Nugent, T. C.; Hudlicky, T. J. Org. Chem. 1998, 63, 510–520.
18. Schwab, J. M.; Lin, D. C. T. J. Am. Chem. Soc 1985, 107,
6046–6052.
Hart, J.; Nelson, A. J. Chem. Soc., Perkin Trans 1 2000,
711–722. To further ascertain the presence of an oxetane (and
not of an oxirane), 7 was treated with isopropyldimethylsilyl
chloride which led to selective silylation of the allylic alcohol
(determined by COSY NMR). N-PSP-induced cyclization
afforded two compounds: 9 and the corresponding silyl ether
(which was identical to the compound obtained by silylation
of 9).
19. Oestreich, M.; Frohlich, R.; Hoppe, D. J. Org. Chem. 1999, 64,
8616–8626.
20. Vargese, C.; Abushanab, E. J. Org. Chem. 1990, 55,
4400–4403.
15. Most of mono-THF-acetogenins shared as a common moiety a
2,5-disubsituted THF flanked by two a-hydroxyls. The relative
position of these structural elements defines a variety of
subfamilies featuring threo-trans-threo, threo-cis-threo, erythro-
trans-threo, threo-trans-erythro, etc. a,a-bis-hydroxy-THF
system. For reviews on acetogenins, see: (a) Alali, F. Q.; Liu,
X.-X.; McLaughlin, J. L. J. Nat. Prod. 1999, 62, 504–540. (b)
Figadere, B. Acc. Chem. Res. 1995, 28, 359–365.
21. An efficient synthesis of both (E) and (Z)-2-ene-1,5-diols has
recently been reported: Flamme, E. M.; Roush, W. R. J. Am.
Chem. Soc. 2002, 124, 13644–13645.
22. Tetrahydrofuran 29 has NMR data identical to those reported
in the literature,2 except for optical rotation ([a]2D0¼þ3.3
(c 0.9, CHCl3). In order to make sure that 29 had the correct
structure, it was converted to the corresponding MOM ether
which had a value similar ([a]2D0¼þ29.0 (c 0.3, CHCl3) to that
reported in Ref. 2: [a]2D0¼þ30.1 (c 0.17, CHCl3)).
16. For a recent approach to the various types of THF cores in