A Method for Syn-Dihydroxylation of Double Bonds Cis to a Hydroxymethyl Substituent

Full text of DOI:10.1021/jo9912808

J . Org. Chem. 2000, 65, 2797-2801
2797
A Meth od for Syn -Dih yd r oxyla tion of
Dou ble Bon d s Cis to a Hyd r oxym eth yl
Su bstitu en t
Rasmus P. Clausen and Mikael Bols*
Department of Chemistry, Aarhus University,
Aarhus DK-8000 C, Denmark
Received August 11, 1999
In tr od u ction
Syn dihydroxylation of olefins is an extremely useful
synthetic transformation in organic chemistry. It is
usually carried out with OsO4 or KMnO4 giving normally
preferential addition to the double bond from the least-
hindered side of the molecule.1 Even though examples
of preferential addition of the diol from the more-hindered
side do exist,2 they appear to be exceptional cases.
Substrate-directed addition, which is successful for many
other reactions, does not work well for osmylations,3 and
therefore control of the diastereoselectivity of the trans-
formation seems to be a problem.
Nevertheless, addition of the diol from the more-
hindered face could be very useful. In a research program
where carbohydrate mimetics with a galactose type
stereochemistry was needed (carbocyclic or heterocyclic
six-ring systems with an overall syn cis 2,3-dihydroxy-
1-hydroxymethyl substitution, Figure 1), it was obvious
that a potentially very convient way of obtaining these
compounds would be by a dihydroxylation of the corre-
sponding homoallylic alcohol from the hydroxymethyl
face. Osmium tetraoxide-catalyzed dihydroxylation can-
not be applied to obtain this stereochemistry: Kishi has
shown that dihydroxylation of 3-methylcyclohexene with
OsO4 gives exclusively addition from the anti face,4 while
similar dihydroxylation of N-methyl-3-hydroxymethyl-
piperid-4-ene gives a 5:1 ratio of anti over syn addition.5
In contrast to dihydroxylation, intramolecular addition
of nucleophiles to an olefin activated by electrophiles
occurs from the syn face with high stereoselectivity.6 The
arch-type reaction of this kind is the well-known halo-
lactonization;7 this procedure has been further developed
to tethered versions involving cyclic carbonates8 and
ureas9 as well as haloetherifications10 and halosiloxyl-
F igu r e 1.
ations.11 These reactions add an oxygen nucleophile from
the syn face and a leaving group from the anti face. Thus
nucleophilic substitution of the leaving group with an
oxygen nucleophile should lead to the desired syn syn
diol. The halocarbonate reactions have been used to
epoxidize the syn face, but there are to our knowledge
no examples of a transformation that results in a syn
dihydroxylation of the hydroxymethyl face of a ring.12
In this paper we present an indirect procedure based
on halolactonization that allows diastereoselective cis/
cis dihydroxylation of a homoallylic double bond. We also
present a new haloacetalization reaction.
Resu lts a n d Discu ssion
To use a halolactonization type reaction for hydroxyl-
directed dihydroxylation a tethered reaction was required
in which the tether could be connected to a hydroxy-group
and contained an O-nucleophile. Two solutions were
found to this problem: A method based on the halolacton-
ization reactions of Rousseau7b and a haloacetalization
reaction using the 2-trimethylsilylethoxymethyl (SEM)
group. In the first case an acetate residue is connected
to the hydroxyl group and this is used to direct attack in
a halolactonization. Later the acetate is removed by
R-bromination and hydrolysis. In the second case an
allylic or homoallylic alcohol is protected as a SEM group,
which then with halonium ions undergo reaction to form
a methylene acetal.
All the halonium-promoted reactions reported in this
paper were attempted with NIS, iodonium dicollidine
perchlorate (IDCP), ICl, and I2/KI as well as iodonium
dicollidine hexafluorophosphate (IDCH) and bromonium
dicollidine hexafluorophosphate (BrDCH), but similarly
to Rousseau7b,c we found that IDCH and BrDCH gave the
best results. Since no proper literature procedure for
synthesis of these reagents have been reported, we
describe their synthesis in the Experimental Section.
(1) For some examples related to those of this work: (a) Mehta, G.;
Mohal, N. Tetrahedron Lett. 1998, 39, 3281-4. (b) Oida, E.; Ohki, E.
Chem. Pharm. Bull. 1969, 17, 1990-1999. (c) Bach, P.; Lohse, A.; Bols,
M. Tetrahedron Lett. 1999, 40, 3461-4. (d) Bols, M.; Hazell, R.;
Thomsen, I. Chem. Eur. J . 1997, 3, 940-947.
(2) (a) Hildbrand, S.; Leumann, C.; Scheffold, R. Helv. Chim. Acta
1996, 79, 702-9. (b) Donohoe, T. J .; Moore, P. R.; Waring, M. J .;
Newcombe, N. J . Tetrahedron Lett. 1997, 38, 5027-30. (c) Donohoe,
T. J .; Blades, K.; Helliwell, M.; Moore, P. R.; Winter, J . J . G.; Stemp,
G. J . Org. Chem. 1999, 64, 2980-1. (d) Harris, J . M.; Keranen, M. D.;
O’Doherty, G. A. J . Org. Chem. 1999, 64, 2982-3.
(3) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. Rev. 1993, 93,
1307-1370.
(4) Cha, J . K.; Christ, W. J .; Kishi, Y. Tetrahedron Lett. 1983, 24,
3943-3946.
(5) Ulldahl, S. U.; Bols, M. Unpublished.
(9) Kohn, H.; Bean, M. B.; Rohrscheidt, C. von; Willcott, M. R.;
Warnhoff, E. W. Tetrahedron 1981, 37, 3195-3201.
(10) Kocovsky, P.; Pour, M. J . Org. Chem. 1990, 55, 5580-5589.
(11) Takaku, K.; Shinokubo, H.; Oshima, K. Tetrahedron Lett. 1981,
37, 3195-3201.
(6) Cardillo, G.; Orena, M. Tetrahedron 1990, 46, 3321-3408.
(7) (a) Dowle, M. D.; Davies, D. I. Chem. Soc. Rev. 1979, 8, 171-
197. (b) Simonot, B.; Rousseau, G. J . Org. Chem. 1993, 58, 4-5. (c)
Homsi, F.; Rousseau, G. J . Org. Chem. 1998, 63, 5255-5258.
(8) Bongini, A.; Cardillo, G.; Orena, M.; Porzi, G.; Sandri, S. J . Org.
Chem. 1982, 47, 4626-4633.
(12) For an example where the cis-cis-2,3-dihydroxy-1-hydroxy-
methyl substructure was obtained in another way, see ref 1a.
10.1021/jo9912808 CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/07/2000

2798 J . Org. Chem., Vol. 65, No. 9, 2000
Notes
Sch em e 1
Sch em e 2
The study was carried out on cyclohex-2-enemethanol
(1), which was synthesized from cyclohexene by depro-
tonation with Schlossers base13 (BuLi/ButK) and alkyl-
ation with paraformaldehyde (Scheme 1). This gave 1 in
65% yield.12
Sch em e 3
Ha lola cton iza tion Str a tegy. The homoallyl alcohol
1 was reacted with chloroacetate in the presence of NaH
and NaI (nucleophilic catalyst) in THF to give the
corresponding carboxymethylated alcohol 2 in 91% yield
(Scheme 1). Treatment of 2 in CH2Cl2 with IDCH, formed
in situ from the silver salt and I2, gave the iodolactone 3
as a single product in 62% yield. Rousseau has carried
out the same type of reaction on acyclic systems, forming
seven-membered lactones in good yield.7b However, the
fact that this bicyclic version of his reaction is stereo-
selective is remarkable, because the lactone is so large
that a trans-fused system could easily be formed. The
structure of 3 was therefore critically analyzed by 1H
NMR. The most downfield proton H-1 (δ 4.95) only has
small couplings. This is inconsistent with trans-fused
rings, because such a stereochemical relationship would
force H-1 and H-7 to be axial giving a large J 1,7. The
substitution at C-1 and C-7 must therefore be cis as
depicted. The very downfield chemical shift of H-1 also
show that it is equatorial. Furthermore, H-11 (δ 4.63)
has relatively small couplings as well (J e 6 Hz) showing
that this proton is equatorial, and the iodine axial. This
means that the substituents at C-1 and C-11 have a trans
relationship. Stereoselectivity is probably caused by
stringent geometrical requirements to the position of the
nucleophile in order to be able to open the iodonium ion.
The carboxylate has to attack from an almost axial
position, and it is difficult to meet these requirements
on the anti face. The reaction was also attempted with
BrDCH, giving the corresponding bromolactone 4 in 28%
yield.
small couplings, while H-11 (δ 3.85) has one large
coupling (J ) 9.5 Hz). This shows that H-11 is axial and
has one diaxial coupling, but not to H-1. Therefore, H-1
must be equatorial, and the C-1/C-11 substituents must
consequently have a cis stereochemistry.
The acetate tether was removed using the following
sequence: Lactone 5 was subjected to radical bromina-
tion conditions using NBS in CCl4 at reflux16 in the
presence of radical initiator lauryl peroxide. This reaction
gave the bromide 6 in 85% yield as one compound but
with unknown stereochemistry at C-4. The bromoether
6 was subjected to hydrolysis using aqueous ammonia
at 0 °C, which removed both bromoacetate and trifluoro-
acetate to give the triol 7. This compound was character-
ized as the triacetate 7a , which was obtained in 70% yield
from 6. The stereochemistry could also be confirmed from
the NMR spectra of 7a . The low field chemical shift of
H-2 (δ 5.32) shows that it is an equatorial proton. The
H-3 proton (δ 4.69) has a large coupling (J ) 9 Hz) to
another proton, which shows that it is axial. Their large
coupling is not to H-2, however, which again shows that
H-2 is equatorial. Thus the acetylated diol has cis
stereochemistry.
The iodide 3 was now substituted with trifluoroacetate
by reaction with AgO2CF3 in EtOAc at 0 °C (Scheme 2).
This gave the trifluoroacetate 5 with inversion of con-
figuration in a remarkable yield of 91%. In a case such
as this, where the reagent is a silver salt, the configu-
rational assignment should be carried out with caution,
as both inversion or retention of configuration could be
anticipated. Indeed, literature examples of both stereo-
chemical behaviors, as well as rearrangements, can be
found for iodide substitution with AgO2CF3.15 However,
Ha loa ceta liza tion Str a tegy. Reaction of 1 with SEM
chloride and Hu¨nigs base in CH2Cl2 gave the SEM ether
8 in 99% yield. This compound was reacted with IDCH,
generated in situ, in CH2Cl2. This gave the iodoacetal 9
1
(15) (a) Bloodworth, A. J .; Bowyer, K. J .; Mitchell, J . C. J . Org.
Chem. 1987, 52, 1124-28. (b) Larock, R. C. J . Org. Chem. 1974, 39,
3721-7. (c) Takahata, H.; Takamatsu, T.; Chen, Y.-S.; Ohkubo, N.;
Yamazaki, T. J . Org. Chem. 1990, 55, 3792-7. (d) Burgey, C. S.;
Vollerthum, R.; Fraser-Reid, B. J . Org. Chem. 1996, 61, 1609-18.
(16) (a) Colin, J . L.; Loubinoux, B. Synthesis 1983, 7, 568-571. (b)
Beckwith, A. L. J .; Chai, C. L. L. Tetrahedron 1993, 49, 7871-7882.
as with the iodide 3, H NMR clearly shows the relative
stereochemistry of 5. The H-1 proton (δ 4.78) only has
(13) Hartmann, J .; Schlosser, M. Helv. Chim. Acta 1976, 59, 453-
466.
(14) Alber, F.; Szeimies, G. Tetrahedron Lett. 1994, 35, 4093-6.

Notes
J . Org. Chem., Vol. 65, No. 9, 2000 2799
Sch em e 4
as the main product in 47% yield. The fluorinated
derivative 10 was isolated as a byproduct in 11% yield.
The regio- and stereochemistry of the two products were
determined from one- and two-dimensional NMR spectra.
In 9 the iodide carbon is again easily assigned in
HETCOR as the upfield 13C (32.7 ppm) correlating to a
downfield 1H (4.52 ppm). In COSY this proton couples
to a proton below 2.0, which is from a CH2 rather than a
CH, meaning that iodide has substituted at C-10, cou-
pling to a proton on C-1 (δ 3.98). As in 3 and 5 the
arguments of cis relationship between C-1 and C-6 holds
in bicyclic iodide 9, too. The H-1 only has small couplings
and must be equatorial. This is the case with the H-10
also, and the configuration of 9 must therefore be as
shown in Scheme 3. In compound 10 the fluorine-
substituted carbon exhibits excessive coupling in 13C
NMR. Only three carbons showed C-F couplings: A peak
at 91.7 ppm shows one-bond coupling, while two signals
at 37.2 and 25.4 ppm showed two-bond couplings. This
leads to the conclusion that the fluoride is substituted
at C-3; otherwise it would not couple with the C-4 (25.4
ppm). The HETCOR of 10 also reveals the C-2 (37.2 ppm)
and H-2 (δ 4.7) to be the iodo-substituted site, and this
carbon couples with fluorine as expected. The H-3 (δ 5.02)
only has small couplings and must therefore be equato-
rial. The H-2 signal has a relatively low chemical shift
of δ 4.7, which also fits an equatorial proton as compared
with H-10 in 9 or H-11 in 3. Thus, the two halogen atoms
must be trans. Assuming a chair conformation with two
trans axial substituents, it is fair to conclude that the
C-1 SEM,oxymethyl-substituent is equatorial since if it
was axial the compound would be expected to shift to the
all-equatorial chair confomation. It is therefore concluded
that the stereochemistry of 10 is as shown in Scheme 3.
The formation of stereoisomer 9 is expected to give the
stereoselectivity of the reaction 2 to 3 since in this case
the new ring is smaller. The formation of the fluoride 10
is more surprising. The stereochemistry of 10 suggests
that it is formed from the cis iodonium ion that cannot
lead directly to 9. In any case the lesser efficiency of this
reaction compared to the halolactonization makes this
strategy less attractive.
the only product. The structure including stereochemistry
is easily seen from the observations obtained by 1H NMR
and HETCOR that the proton (H-5) of the iodo-substi-
tuted carbon has three couplings of which two are large.
This is only possible in the isomer shown (Scheme 4).
Reaction of 11 with bromonium dicollidine hexafluoro-
phosphate (BrDCH) was an even faster and more-
efficient reaction, giving the corresponding bromide 13
in 74%. The reaction occurred with rapid evolution of a
colorless gas presumably ethylene. Gas evolution was
also observed in the formation of 12 and 9 but at a much
slower pace. The structure of 13 was determined simi-
larly to 12 from the couplings of H-5 (δ 4.14).
The reaction also worked on the methyl analogue 15
(obtained by SEM protection of 14 in 99% yield). Treat-
ment with of 15 with BrDCH gave the dioxane 16 in 62%
yield as the only product. The different regioselectivity
on the reations of cinnamoyl derivatives, as compared
with the cyclohexene system, leading to endo and not exo
opening of the halonium ions, is undoubtedly caused by
the stabilizing effect of the phenyl group toward carbo-
cations. The results show that the haloacetalization
reaction of a SEM group is an interesting reaction but
less effective for solving the problem at hand.
The present work has shown that carboxymethylation
of a homoallylic alcohol with chloroacetic acid followed
by IDCH-promoted iodolactonization and substitution of
the iodide with silver trifluoroacetate can be used for
diastereoselective syn-syn dihydroxylation of the double
bond. The method works on a cyclic alkene. It is likely
that the method can be extended to cyclic allylic alcohols
as well since the basic principles governing the stereo-
selectivity should be the same.
Exp er im en ta l Section
Gen er a l. 13C NMR and 1H NMR spectra were recorded on a
Varian Gemini 2000 instrument using DHO (1H NMR: δ 4.7
ppm) or acetone (1H NMR: δ 2.05 ppm; 13C NMR: δ 29.8 ppm)
as reference. EI mass spectra were obtained using a VG TRIO-2
spectrometer. Concentrations were carried out on a rotary
evaporator at temperatures below 40 °C. TLC was performed
on Kieselgel 60 F254 glass-backed plates (Merck) with detection
by soaking in 1% cerium sulfate and 1.5% molybdic acid in 10%
aqueous H2SO4 followed by heating to 200 °C for 60 s. The
pyridine was dried over KOH.
We also tried this new haloacetalization reaction on
allyl alcohols. The SEM-protected alcohol 11 was syn-
thesized by reaction of cinnamoyl alcohol with SEM
chloride and Hu¨nigs base (Scheme 4). Reaction of 11 with
IDCH in CH2Cl2 gave the iodo acetal 12 in 62% yield as
Silver Dicollid in e Hexa flu or op h osp h a te. AgNO3 (9.00 g,
53.0 mmol) and NaPF6 (11.0 g, 65.5 mmol) were dissolved in
H2O (100 mL), and 2,4,6-collidine (20.0 mL) was added. After

2800 J . Org. Chem., Vol. 65, No. 9, 2000
Notes
stirring for 30 min. the precipitate was filtered on a sintered
glass funnel and washed with H2O (4×, EtOH (1×), and Et2O
(2×). The white material was dried in a desiccator in vacuo over
silica, giving the title compound (25.9 g, 52.3 mmol, 99%).
Br om on iu m Dicollidin e Hexaflu or oph osph ate (Br DCH).
To silver dicollidine hexafluorophosphate (9.90 g, 20.0 mmol) in
CH2Cl2 (100 mL) was added Br2 (∼1.00 mL, ∼3.20 g, 20.0 mmol)
dropwise. Addition was continued until the solution became
colored. The solution was filtered through Celite, leaving the
precipitated AgBr on the filter. Upon evaporation, Et2O was
added to the residue. The precipitate was filtered, washed with
pentane, and dried in vacuo, giving bromonium dicollidine
hexafluorophosphate (7.84 g, 16.8 mmol, 84%) as off-white
crystals.
Iod on iu m Dicollid in e Hexa flu or op h osp h a te (IDCH). To
silver dicollidine hexafluorophosphate (13.8 g, 27.9 mmol) in
CH2Cl2 (150 mL) was added finely ground I2 (7.00 g). Pale yellow
AgI precipitated and after 30 min the solution was filtered
through Celite. The solution was evaporated to ∼20 mL, and
Et2O (50 mL) was added. Filtration, washing with pentane, and
drying in vacuo gave the title compound as off-white crystals
(12.1 g, 23.5 mmol, 85%).
Cycloh ex-2-en em eth a n ol (1). Potasium tert-butoxide (17.0
g, 0.152 mol) was suspended in distilled cyclohexene (110 mL)
degassed with N2. BuLi (100 mL, 1.6 M, 0.16 mol) was added
dropwise over 2 h, keeping the temperature under 40 °C. The
reaction was left stirring overnight leaving a brownish yellow
suspension. The mixture was heated to 60 °C. Paraformaldehyde
(5.2 g, 0.172 mol) (dried over silica) was carefully added
portionwise causing an exothermic reaction upon every addition.
At the end of addition a clear light green solution evolves which
was left stirring for 1 h. The solution was cooled and poured
into NH4Cl (aq) at 0 °C. The organic phase was separated and
the aqueous phase extracted with 3 × 150 mL of CH2Cl2. The
collected organic phases were washed with 3 × 100 mL H2O/
NaCl(aq)/H2O, dried (MgSO4), and evaporated. Distillation gave
11.0 g of 1 (0.098 mol, 65%).14 1H NMR (200 MHz, CDCl3): δ
5.6-5.5 (dm, 1H), 5.45-5.35 (dm, 1H), 3.45 (s, 1H), 3.25 (d, 2H),
2.15-2.0 (m, 1H), 1.85-1.75 (m, 2H), 1.65-1.5 (m, 2H), 1.45-
1.3 (m, 1H), 1.25-1.1 (m, 1H). 13C NMR (50 MHz, CDCl3): δ
128.7, 127.8, 66.4, 37.9, 25.3, 25.1, 20.7.
Cycloh ex-2-en ylm eth oxya cetic Acid (2). To 2.24 g (20.0
mmol) of 1, 2.1 g (22 mmol) of chloroacetic acid, and a catalytic
amount of NaI in 50 mL of THF was added 2.4 g (80 mmol) NaH
(80% dispersion in mineral oil). The mixture was refluxed
overnight and poured on 100 mL H2O (0 °C). The aqueous basic
solution was washed with 100 mL of Et2O, acidified (H2SO4),
and extracted with 3 × 100 mL of CH2Cl2. The collected organic
phases were dried (MgSO4) and evaporated. Column chroma-
tography (SiO2, 2% AcOH in EtOAc) gave 3.10 g (18 mmol, 91%)
of 2. 1H NMR (200 MHz, CDCl3): δ 11.1 (s, 1H) 5.6-5.5 (dm,
1H), 5.45-5.35 (dm, 1H), 3.95 (s, 2H), 3.25 (d, 2H), 2.25-2.15
(m, 1H), 1.8-1.7 (m, 2H), 1.7-1.05 (m, 4H). 13C NMR (50 MHz,
CDCl3): δ 175.7, 128.9, 127.4, 75.9, 67.7, 35.4, 25.6, 25.1, 20.6.
(1RS,7SR,11RS)-2,5-Dioxa -11-iod o-3-oxob icyclo[5.4.0]-
u n d eca n e (3). To 9.90 g (20.0 mmol) of silver dicollidine
hexafluorophosphate dissolved in 150 mL of CH2Cl2 was added
5.10 g (20.0 mol) finely grounded I2. The solution was filtered
through oven-dried Celite and added to 2.0 g (11.8 mmol) of 2.
The mixture was refluxed for 4 h. The resulting solution was
washed with 5% Na2S2O3 (aq) and evaporated. Column chro-
matography (SiO2/CH2Cl2) gave 2.15 g (7.30 mmol, 62%) of 3.
1H NMR (200 MHz, CDCl3): δ 4.95 (bs, 1H, H-1), 4.63 (dd, 1H,
J ) 6 and 3 Hz, H-11), 4.41 (d, 1H, J ) 16 Hz, H-4a), 4.26 (d,
1H, J ) 16 Hz, H-4b), 3.89 (dd, 1H, J ) 5.5 and 12.5 Hz, H-6a),
3.49 (dd, 1H J ) 5.5 and 12.5 Hz, H-6b), 2.85-2.71 (m, 1H, H-7),
2.04-1.93 (m, 1H, H-10a), 1.87-1.40 (m, 5H). 13C NMR (50 MHz,
CDCl3): δ 169.7 (C-3), 78.4 (C-1), 74.8 (C-6), 71.9 (C-4), 35.9 (C-
7), 29.4 (C-11), 29.2, 23.7, 21.1. (Assignments based on COSY
and HETCOR spectra). MS (EI): m/z 297(M+ + 1).
) 17 Hz, H-4a), 4.28 (d, 1H, J ) 17 Hz, H-4b), 3.93 (dd, 1H, J
) 5.5 and 12.5 Hz, H-6a), 3.43 (dd, 1H, J ) 5.5 and 12.5 Hz,
H-6b), 2.73-2.59 (m, 1H, H-7), 2.33-2.14 (m, 1H, H-10a), 1.97-
1.43 (m, 5H). 13C NMR (50 MHz, CDCl3): δ 169.1, 76.5, 74.1,
70.9, 48.7, 34.6, 26.9, 22.4, 18.5. (Assignments based on COSY
spectra).
(1RS,7SR,11SR)-2,5-Dioxa -11-tr iflu or oa cetoxy-3-oxobi-
cyclo[5.4.0]u n d eca n e (5). To 1.10 g (3.7 mmol) of 3 in 10 mL
of EtOAc was added 1.00 g (4.3 mmol) AgO2CCF3 at 0 °C, and
the mixture was left stirring for 4 h at room temperature. Then
10.0 mL of pentane was added, and the suspension was filtered
through a pad of SiO2 (1:1 EtOAc/pentane). Evaporation gave
0.95 g (3.40 mmol, 91%) of 5. 1H NMR (200 MHz, CDCl3): δ
4.78 (t, 1H, J ) 3 Hz, H-1), 4.41 (d, 1H, J ) 2 Hz, H-6a), 4.37 (s,
1H, H-6b), 4.33 (s, 1H, H-4a), 4.32 (s, 1H, H-4b), 3.85 (ddd, 1H,
J ) 3, 6.5 and 9.5 Hz, H-11), 2.25-2.05 (m, 1H, H-7), 1.95-1.62
(m, 3H) 1.59-1.27 (m, 3H). 13C NMR (50 MHz, CDCl3): δ 167.4
(C-3), 157.1 (q, J ) 43 Hz, CF3CO), 114.1 (q, J ) 285 Hz, CF3),
75.3 (C-1), 70.1 (C-11), 66.7 (C-6), 61.1 (C-4), 38.9 (C-7), 23.7
(C-10), 21.8 (C-9), 21.5 (C-8). (Assignments based on COSY and
HETCOR spectra). MS (EI): m/z 282 (M+).
(1RS,7SR,11SR)-4-Br om o-2,5-d ioxa -11-tr iflu or oa cetoxy-
3-oxobicyclo[5.4.0]u n d eca n e (6). To 1.20 g (4.25 mmol) of 5
in 25.0 mL of CCl4 was added 0.85 g (4.78 mmol) N-bromosuc-
cinimide and a spatula point of lauryl peroxide. The solution
was refluxed for 30 min. Filtration and evaporation gave 1.30 g
(3.60 mmol, 85%) of crude 6. Due to instability the product was
1
not further purified. H NMR (200 MHz, CDCl3): δ 6.41 (s, 1H,
H-4), 5.53 (t, 1H, J < 2 Hz, H-1), 4.41 (d, 1H, J ) 2 Hz, H-6a),
4.37 (s, 1H, H-6b), 4.23-4.12 (m, 1H, H-11), 2.20-2.07 (m, 1H,
H-7), 1.89-1.78 (m, 2H), 1.63-1.30 (m, 4H). 13C NMR (50 MHz,
CDCl3): δ 161.6 (C-3), 157.0 (q, J ) 42 Hz, CF3CO), 114.3 (q, J
) 286 Hz, CF3), 73.9 (C-4), 73.0 (C-11), 71.5 (C-1), 66.5 (C-6),
38.0 (C-7), 27.0, 21.5, 21.4. (Assignments based on COSY and
HETCOR spectra). MS (EI): m/z 363(M+ + 3), 361(M+ + 1).
(1S R ,2R S ,3S R )-2,3-D ia c e t o x y c y c lo h e x a n e m e t h a n y l
Aceta te (7a ). 6 (1.30 g, 3.60 mmol) was treated with NH3
(aq, 25%, 25 mL) at 0 °C. Cooling was removed, and the mixture
was left stirring at 25 °C until finished (TLC, 10% MeOH in
EtOAc). The solution was evaporated, and the residue was
suspended in MeOH. The suspension was filtered through a pad
of SiO2 (10% MeOH in EtOAc) and evaporated, giving 610 mg
of crude triol 7. (1H NMR (300 MHz, D2O): δ 3.86 (bs, 1H, H-2),
3.52 (dd, 1H, J ) 11, 7.5 Hz, H-1′a), 3.49 (ddd, 1H, J ) 11, 4.5
and 2.5 Hz, H-3), 3.39 (dd, 1H, J ) 11, 6.5 Hz, H-1′b), 1.70-
0.97 (m, 7H). 13C NMR (75 MHz, D2O): δ 72.4 (C-3, 70.1 (C-2),
63.8 (C-1′), 42.8 (C-1), 27.7, 23.0, 21.7 (Assignments based on
COSY and HETCOR spectra). MS (EI): m/z 146 (M+). Peak
match:146.0943 (calcd:146.0943).) This sample was treated with
pyridine and acetic anhydride. The solution was evaporated and
water was added. Extraction (CH2Cl2), evaporation, and column
chromatography gave 680 mg (2.50 mmol, 70% from 6) of 7a as
an oil. 1H NMR (200 MHz, CDCl3): δ 5.32 (t, 1H, J ) 2.5 Hz,
H-2), 4.69 (ddd, 1H, J ) 9, 7 and 2.5 Hz, H-3), 3.85 (dd, 1H, J )
11 and 9 Hz, H-1′a), 3.76 (dd, 1H, J ) 11 and 6.5 Hz, H-1′b),
2.00 (s, 3H, Ac), 1.93 (s, 3H, Ac), 1.89 (s, 3H, Ac), 1.90-1.16 (m,
7H). 13C NMR (50 MHz, CDCl3): δ 170.7 (Ac), 170.1 (Ac), 170.0
(Ac), 72.2 (C-3), 67.9 (C-2), 63.7 (C-1′), 38.5 (C-1), 25.4, 22.5, 22.3,
20.8 (Ac), 20.64 (Ac), 20.61 (Ac). (Assignments based on COSY
and HETCOR spectra). MS (EI): m/z 272 (M+).
C y c lo h e x -2-e n y lm e t h o x y (2-t r im e t h y ls ily le t h o x y )-
m eth a n e (8). To 1.12 g (10.0 mmol) of 1 in 10.0 mL CH2Cl2
were added 1.85 g of SEM-Cl (11.1 mmol) and 2 mL of DIPEA
(11.5 mmol). When the reaction was complete (TLC, CH2Cl2, 1
h), the solvent was evaporated. Column chromatography (SiO2/
1
CH2Cl2) gave 2.39 g (10.0 mmol, 99%) of 8. H NMR (200 MHz,
CDCl3): δ 5.74 (ddd, 1H, J ) 3.5, 6.0 and 10.0 Hz), 5.57 (ddd,
1H, J ) 1.8, 4.2 and 10.0 Hz) 4.65 (s, 2H) 3.60 (t, 2H, J ) 8.4
Hz), 3.98 (d, 2H, J ) 6.4 Hz), 2.43-2.22 (m, 1H), 2.02-1.93 (m,
2H) 1.84-1.12 (m, 4H), 0.92 (t, 2H, J ) 8.4 Hz), 0.00 (s, 9H).
13C NMR (50 MHz, CDCl3): δ 128.6, 128.2, 94.8, 71.8, 64.8, 35.8,
26.0, 25.2, 20.9, 18.0, -1.4.
(2RS,4SR,10RS)-2,4-Dioxa-10-iodobicyclo[4.4.0]decan e (9)
a n d (1SR,2RS,3RS)-3-F lu or o-2-iod ocycloh exa n ylm eth oxy-
(2-tr im eth ylsilyleth oxy)m eth a n e (10). To 2.00 g (4.04 mmol)
of silver dicollidine hexafluorophosphate dissolved in 40.0 mL
of CH2Cl2 was added 1.03 g (4.06 mmol) of finely grounded I2.
(1RS,7SR,11RS)-11-Br om o-2,5-d ioxa -3-oxobicyclo[5.4.0]-
u n d eca n e (4). To 170 mg (1.00 mmol) of 2 in 15.0 mL of CH2Cl2
was added 1.00 g (2.14 mmol) of bromonium dicollidine hexa-
fluorophosphate. The mixture was refluxed for 4 h. Evaporation
and column chromatography (SiO2, EtOAc/pentane (1:1)) gave
70 mg of the bromide 4 (0.28 mmol, 28%). 1H NMR (200 MHz,
CDCl3): δ 4.92 (bs, 1H, H-1), 4.50 (m, 1H, H-11) 4.47 (d, 1H, J

Notes
J . Org. Chem., Vol. 65, No. 9, 2000 2801
The solution was filtered through oven-dried Celite and added
to 485 mg (2.00 mmol) of 8 and left stirring overnight. The
solution was washed with Na2S2O3 (5% aq) and evaporated.
Column chromatography gave 250 mg (0.93 mmol, 47%) of 9 and
85 mg of a faster running compound which was identified to be
10 (11%). 9: 1H NMR (200 MHz, CDCl3): δ 4.97 (d, 1H, J ) 6.5
Hz, H-3a), 4.64 (d, 1H, J ) 6.5 Hz, H-3b), 4.52 (dd, 1H, J ) 5
and < 2 Hz, H-10), 3.98 (d, 1H, J < 2 Hz, H-1), 3.78 (d, 2H, J <
2 Hz, H-5′), 2.12-1.40 (m, 7H). 13C NMR (50 MHz, CDCl3): δ
93.9 (C-3), 78.5 (C-1), 72.1 (C-5), 32.7 (C-10), 31.1 (C-6), 29.6,
23.9, 21.5. (Assignments based on COSY and HETCOR spectra).
MS (EI): m/z 268 (M+). 10: 1H NMR (200 MHz, CDCl3): δ 5.02
(d, 1H, J ) 48.2 Hz, H-3), 4.7 (m, 3H, H-2, OCH2O), 3.59 (t, 2H,
J ) 8 Hz), 3.44-3.23 (m, 2H, H-1′a, H-1′b), 2.54-2.12 (dm, 1H,
J ) 45 Hz), 2.0-1.15 (m, 6H), 0.93 (t, 2H, J ) 8 Hz), 0.00 (s,
9H). 13C NMR (50 MHz, CDCl3): δ 95.0 (OCH2O), 91.7 (d, J )
176 Hz, C-3), 73.0 (C-1′), 65.2, 37.2 (d, J ) 23 Hz, C-2), 36.1,
25.4 (d, J ) 21 Hz, C-4), 24.2, 19.6, 18.1, -1.4. (Assignments
based on COSY and HETCOR spectra). MS (EI): m/z 387 (M+
- H, vw).
1-O-[(2-Tr im eth ylsilyleth oxy)m eth yl]cin n a m yl Alcoh ol
(11). To 1.34 g (10 mmol) cinnamyl alcohol in 10.0 mL of CH2Cl2
were added 2.00 mL of DIPEA and 2 mL (1.88 g, 11.3 mmol) of
SEMCl. After 4 h the solution was evaporated, and column
chromatography (SiO2, Et2O/pentane, 1:4) gave 2.32 g (8.8 mmol,
88%) of 11. 1H NMR (200 MHz, CDCl3): δ 7.43-7.18 (m, 5H),
6.63 (d, 1H, J ) 16 Hz), 6.31 (dt, 1H, J ) 16, 6 Hz), 4.76 (s, 2H),
4.26 (d, 2H, J ) 6 Hz), 3.68 (t, 2H, J ) 8 Hz), 0.95 (t, 2H, J )
8 Hz), 0.00 (s, 9H). 13C NMR (50 MHz, CDCl3): δ 136.6, 132.4,
128.5, 127.6, 126.4, 125.6, 92.9, 67.8, 65.1, 18.0, -1.4.
(4,5-tr a n s)-5-Iod o-4-p h en yl-1,3-d ioxa n e (12). To 265 mg
(1.00 mmol) of 11 in 5.00 mL of CH2Cl2 was added 1.0 g (1.90
mmol) IDCH, and the mixture was left stirring for 4 h at 25 °C.
Washing with Na2S2O3 (5% aq) followed by column chromatog-
raphy (SiO2, CH2Cl2) gave 180 mg (0.62 mmol, 62%) of 12. 1H
NMR (200 MHz, CDCl3): δ 7.36-7.32 (m, 5H, Ph), 5.27 (dd*,
1H, J ) 6.5 Hz, H-2eq), 4.90 (d, 1H, J ) 6.5 Hz, H-2ax), 4.58 (d,
1H, J ) 10 Hz, H-4), 4.39 (ddd*, 1H, J ) 11, 4.5 Hz, H-6eq),
4.25 (ddd, 1H, J ) 11, 10 and 4.5 Hz, H-5), 3.92 (t, 1H, J ) 11
Hz, H-6ax). * very small long-range coupling due to w-configu-
ration can be observed. 13C NMR (50 MHz, CDCl3): δ 138.1,
128.9, 128.3, 127.6, 94.6 (C-2), 85.3 (C-4), 74.0 (C-6), 28.4 (C-5).
(Assignments based on COSY and HETCOR spectra). MS (EI):
m/z 290 (M+). Peak match 289.9805 (calcd 289.9804).
(4,5-tr a n s)-5-Br om o-4-p h en yl-1,3-d ioxa n e (13). To 500 mg
(1.10 mmol) of BrDCH in 2 mL of CH2Cl2 was added 264 mg
(1.00 mmol) of 11. Gas evolved and after 10 min, the solution
was concentrated and subjected to column chromatography
(SiO2, CH2Cl2), which gave 180 mg (0.74 mmol, 74%) of 13. 1H
NMR (200 MHz, CDCl3): δ 7.36-7.32 (m, 5H, Ph), 5.26 (dd*,
1H, J ) 6.5 Hz, H-2eq), 4.91 (d, 1H, J ) 6.5 Hz, H-2ax), 4.56 (d,
1H, J ) 10 Hz, H-4), 4.43 (ddd*, 1H, J ) 11 and 4.5 Hz, H-6ax),
4.14 (ddd, 1H, J ) 10, 11 and 4.5 Hz, H-5), 3.87 (t, 1H, J ) 11
Hz, H-6eq) * very small long-range coupling due to w-configu-
ration can be observed. 13C NMR (50 MHz, CDCl3): δ 137.4,
128.9, 128.3, 127.5, 94.3 (C-2), 84.5 (C-4), 72.1 (C-6), 46.8 (C-5).
(Assignments based on COSY and HETCOR spectra).
(()-1-P h en ylbu t-1-en -3-ol (14). (E)-4-Phenylbut-3-en-2-one
(2.92 g, 0.02 mol) was dissolved in 100 mL of MeOH, and 4.4 g
(0.04 mol) of anhydrous CaCl2 was added. After stirring 30 min,
the solution was cooled to 0 °C, and 1.14 g (0.03 mol) NaBH4
was added portionwise. The solution was left stirring 1 h at 25
°C and poured onto 150 mL of brine. This was extracted twice
with Et2O, dried (MgSO4), and evaporated giving 2.0 g (13.5
mmol; 67%) of 14. 1H NMR (200 MHz, CDCl3): δ 7.27-7.13 (m,
5H), 6.42 (d, 1H, J ) 16 Hz), 6.12 (dd, 1H, J ) 16, 6.2 Hz), 4.34
(dp, 1H, J ) 1.1, 6.2 Hz), 2.46 (s, 1H), 1.24 (d, 3H, J ) 6.2 Hz).
13C NMR (50 MHz, CDCl3): δ 136.6, 133.5, 129.1, 128.4, 127.4,
126.3, 68.6, 23.3.
(()-1-O-[(2-Tr im eth ylsilyleth oxy)m eth yl]-1-p h en ylbu t-
1-en -3-ol (15). To 740 mg (5.00 mmol) of 14 in 5.00 mL of CH2Cl2
was added 1.00 mL of DIPEA and 1.00 mL (0.94 g, 5.60 mmol)
of SEM-Cl. The solution was left stirring until finished (TLC,
EtOAc:pentane 1:4, 4 h). Evaporation and column chromatog-
raphy (SiO2, EtOAc/pentane (1:4)) gave 1.38 g (4.95 mmol, 99%)
of 15. 1H NMR (200 MHz, CDCl3): δ 7.37-7.18 (m, 5H), 6.52
(d, 1H, J ) 16 Hz), 6.06 (dd, 1H, J ) 16, 8 Hz), 4.74 (d, 1H, 7
Hz), 4.66 (d, 1H, J ) 7 Hz), 4.34 (p, 1H, J ) 8 Hz), 3.73 (dt, 1H,
J ) 8, 9 Hz), 3.56 (dt, 1H, J ) 8, 9 Hz), 1.34 (d, 3H, J ) 8 Hz),
0.93 (t, 2H, J ) 8 Hz), 0.00 (s, 9H). 13C NMR (50 MHz, CDCl3):
δ 136.5, 131.3, 130.9, 128.5, 127.5, 126.4, 92.0, 72.4, 64.9, 21.5,
18.1, -1.5.
(4,5-tr a n s-5,6-tr a n s)-5-Br om o-6-m eth yl-4-p h en yl-1,3-d i-
oxa n e (16). To 15 (278 mg, 1.00 mmol) in CH2Cl2 (5.00 mL)
was added BrDCH (0.60 g, 1.32 mmol), and the mixture was
left stirring for 30 min. Evaporation followed by chromatography
(SiO2/CH2Cl2) gave 16 (180 mg, 0.70 mmol, 62%). 1H NMR (200
MHz, CDCl3): δ 7.38-7.32 (m, 5H), 5.18 (d, 1H, J ) 7 Hz), 4.88
(d, 1H, J ) 7 Hz), 4.53 (d, 1H, J 10 Hz), 3.83 (dd, 1H, J ) 11
and 7 Hz), 3.73 (dd, 1H, J ) 10 and 11 Hz), 1.24 (d, 3H, J ) 7
Hz). 13C NMR (50 MHz, CDCl3): δ 137.8, 128.8, 128.3, 127.7,
93.8, 84.3, 78.3, 54.6, 19.7.
Ack n ow led gm en t. We thank the Danish Natural
Research council for financial support. We also thank
B. O. Pedersen for mass spectra.
Su p p or tin g In for m a tion Ava ila ble: Copies of NMR
spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O9912808