The First Tri- and Tetraalkoxysilanes with Four Different Substituents

Article abstract of DOI:10.1021/jo970317q

Unsymmetrically substituted tri- and tetraalkoxysilanes were surprisingly found to be configurationally stable and easy to prepare. It was found that compounds of type MeSi(OR)₂(OR′), MeSi- (OR)(OR′)(OR″), Si(OR)₂(OR′)₂, Si(OR)₂(OR′)(OR″), Si(OR)₃(OR′), and even Si(OR)(OR′)(OR″)(OR*) could be obtained by sequential addition of a variety of chiral and achiral alcohols to methyltrichlorosilane or tetrachlorosilane in the presence of pyridine. In the case of MeSi(OR)(OR′)(OR″) and Si(OR)(OR′)(OR″)(OR*), two types of compounds that have never been prepared before, were obtained in good yield.

Full text of DOI:10.1021/jo970317q

J . Org. Chem. 1997, 62, 4457-4464
4457
Th e F ir st Tr i- a n d Tetr a a lk oxysila n es w ith F ou r Differ en t
Su bstitu en ts
Rasmus P. Clausen and Mikael Bols*
Department of Chemistry, Aarhus University, DK-8000 Aarhus, Denmark
Received February 19, 1997X
Unsymmetrically substituted tri- and tetraalkoxysilanes were surprisingly found to be configura-
tionally stable and easy to prepare. It was found that compounds of type MeSi(OR)2(OR′), MeSi-
(OR)(OR′)(OR′′), Si(OR)2(OR′)2, Si(OR)2(OR′)(OR′′), Si(OR)3(OR′), and even Si(OR)(OR′)(OR′′)(OR*)
could be obtained by sequential addition of a variety of chiral and achiral alcohols to methyltrichlo-
rosilane or tetrachlorosilane in the presence of pyridine. In the case of MeSi(OR)(OR′)(OR′′) and
Si(OR)(OR′)(OR′′)(OR*), two types of compounds that have never been prepared before, were
obtained in good yield.
In tr od u ction
Silyl ethers have found great utility in organic syn-
thesis not only as protecting groups1a but also as reagent
in many different reactions.1b A particularly interesting
use of silicon has been as a tethering agent to achieve
various kinds of selectivity in organic synthesis.2 How-
ever it has mainly been monoalkoxysilanes that have
been employed in the various applications, while dial-
koxysilanes have been much less used, and tri- and
tetraalkoxysilanes not at all. This is intriguing because
the synthetic potential of chiral tri- and tetraalkoxysi-
lanes is tremendous. They might be employed as chiral
tethers, chiral protection groups, or reagents. A chiral
tether would allow not only regio- and diastereoselec-
tivities but also enantioselectivity in the product after
removal of the tether. Particularly compounds that could
be made from tetrachlorosilane (8) or alkyltrichlorosi-
lanes such as methyltrichlorosilane (1) and optically
active alcohols from the chiral pool would be readily
accesible and thus potentially be of great use.
Oligoalkoxysilanes have generally been known to be
configurationally unstable and succeptible to hydrolysis
and disproportionation. (Thus for example, Si(OMe)4
reacts with water in an exothermic reaction.3 ) This view
was supported by a search of the literature, which
revealed that only symmetrical compounds of structure
MeSi(OR)3 and Si(OR)4 were known with very few
exceptions.4,5 Most tri- and tetraalkoxysilanes known
have been made with R being small aliphatic alkyl groups
allowing products to be be isolated by distillation. Only
one example with chiral alkoxygroups was known. Prey
and Gump synthesized various symmetrically substituted
F igu r e 1.
alkoxysilanes of chiral alcohols. The most complex of
these compounds were tri- and tetraalkoxysilanes with
the alkoxy group of a carbohydrate derivative (Figure 1).6
These compounds were obtained in quite low yields
(shown in Figure 1).
In contrast, our experiences with dialkoxydimethylsi-
lanes with bulky alkoxy groups had shown us that these
compounds were rather stable compounds that could be
subjected to chromatography.7 We therefore have en-
deavored to investigate whether unsymmetrical alkoxy-
silanes could be made in good yields in a controlled
manner and whether they would be configurationally or
hydrolytically unstable or undergo spontaneous dispro-
portionation. In this paper we report our findings, which
surprisingly show that these compounds are completely
stable and that it is even possible to prepare a tet-
raalkoxysilane with four different alkoxygroups. We
X Abstract published in Advance ACS Abstracts, J une 1, 1997.
(1) (a) Weber, W. P. Silicon Reagents for Organic Synthesis;
Springer-Verlag: Berlin, 1983. (b) Kociensky, P. J . Protecting groups;
Thieme: Stuttgart, 1994.
(2) Bols, M.; Skrydstrup, T. Chem. Rev. 1995, 95, 1253-1277.
(3) Gmelins handbook of inorganic chemistry, 8th ed; Verlag Chemie
GMBH: Weinheim; Vol. 15c, pp 329-332.
(4) (a) Byk-Guldenwerke. German Patent DE 662732, 1933. (b)
Cella, J . A.; Cargioli, J . D.; Williams, E. A. J . Organomet. Chem. 1980,
186, 13-17.
(5) (a) Onaka, M.; Higuchi, K.; Nanami, H.; Izumi, Y. Bull. Chem.
Soc. J pn. 1993, 66, 2638-45. (b) Matsumoto, H.; Hoshino, Y.; Nagai,
Y. Bull. Chem. Soc. J pn. 1981, 54, 1279-80. (c) Sakurai, H.; Ando,
M.; Kawada, N. Sato, K.; Hosomi, A. Tetrahedron Lett. 1986, 27, 75-
76. (d) Schraml, J .; Krapivin, A. M.; Luzin, A. P.; Kilesso, V. M.;
Pestunovich, V. A. Collect. Czech. Chem. Commun. 1984, 49, 2897-
2902. (e) Sekine, H. Chem. Lett. 1979, 801. (f) Lappert, M. F.; Nile, T.
A. J . Organomet. Chem. 1975, 102, 543-550.
(6) Prey, V.; Gump, K. -H. Liebigs Ann. Chem. 1965, 682, 228-241.
(7) Bols, M. Tetrahedron 1993, 49, 10049-10060 and references
therein.
S0022-3263(97)00317-4 CCC: $14.00 © 1997 American Chemical Society

4458 J . Org. Chem., Vol. 62, No. 13, 1997
Clausen and Bols
Sch em e 1
Sch em e 2
Sch em e 3
demonstrate that simple chiral silyl protection groups can
be created, which can be used to determine the ee of an
alcohol.
Resu lts a n d Discu ssion
We have undertaken a systematic investigation and,
from 1 or 8, attempted to synthesize all the possible types
of unsymmetrically substituted alkoxysilanes (type
1-6): MeSi(OR)2(OR), MeSi(OR)(OR′)(OR′′), Si(OR)2-
(OR′)2, Si(OR)2(OR′)(OR′′), Si(OR)3(OR′), and Si(OR)(OR′)-
(OR′′)(OR*). Compounds of type 1, 2, 4, and 6 have to
our knowledge not previously been prepared. A few
examples of type 34 and type 5 are known.5 In the study
we have employed chiral alcohols that could function as
chiral auxillaries.
Com p ou n d s of Typ e MeSi(OR)3. Initially we in-
vestigated a symmetrically substituted compound similar
to those of Prey and Gump,6 by reacting 1 molar equiv
of 1 with 3.3 molar equiv of 1,2:4,5-diisopropylidene-D-
fructopyranose (2) in pyridine (Scheme 1). This gave the
trialkoxysilane 3 as the only observed product in 70%
yield. This experiment showed two things. A trialkoxy-
silane like 3 was completely stable on TLC or flash
chromatography, which was essential if unsymmetrical
alkoxysilanes were to be made. Second, as our investiga-
tions with 5 (see also below) suggested that 5 substituted
more readily on silicon than 2, the low yield of methyltris-
(1,2:5,6-diisopropylidene-D-glucopyranyloxy)silane ob-
tained by Prey and Gump must have been a result of the
particulars of those reaction conditions. (Only a catalytic
amount of pyridine was used, which may not have been
enough to allow the reaction to run to completion.
Furthermore, purification was carried out by a compli-
cated series of sublimation, extraction, and distillation.6)
Com p ou n d s of Typ e MeSi(OR)2(OR′). From TLC
it became clear that 1 reacted with 2, in the presence of
pyridine, in a sequential manner, so that the silyl chloride
MeSi(OR)2Cl (observed on TLC) could be formed with
selectivity. When 1 and 2 molar equiv of 2 were reacted,
and 1 molar equiv of 2-propanol was added later, mainly
one product, 4, was observed and isolated in 70% yield
by flash chromatography (Scheme 2). Small amounts of
the diisopropoxymonofructose derivative 4a and 3 (7%
each) were also isolated. This showed a high degree
though not complete chemoselectivity in the substitution
of the chlorides at silicon. The two fructose derivatives
of 4 were nonidentical because of the prochiral nature of
the silicon atom, and this could be observed by two
different sets of signals of the fructose moities in the
NMR spectrum. Similarly the spectrum of 4a showed
that the isopropoxy groups gave two sets of signals. This
was interesting because it suggested that the fructose
moiety had a considerable spacial influence on other
substituents, which would be important in any applica-
tion where stereoinduction was desired.
We also attempted sequential reaction of 1 with 2
molar equiv of 1,2:5,6-diisopropylidene-R-D-glucofuranose
(5) and 1 molar equiv of prop-2-enol. In this case the
reaction was less selective and the desired monopro-
penoxydiglucofuranose derivative 6 was obtained only in
41-42% yield after flash chromatography (Scheme 3). In
addition a rather large amount (19%) of the triglucofura-
nose alkoxy-silane 6a was obtained. Molecular models
suggested that the steric hindrance around the silicon
was reduced with 5 as a substituent compared to 2. This
lack of sterical hindrance probably made the sequential
substitution less selective, by making the dialkoxychlo-
rosilane more reactive than that in the above case. The
fact that synthesis of this type of unsymmetrically
substituted alkoxysilane was possible was particularly
important because it opens wide opportunities for use in
tethered reactions. When R* and R′ in trialkoxysilane
R*Si(OR)2(OR′) could react and R was chiral, a poten-
tially enantioselective reaction would occur.
Com p ou n d s of Typ e MeSi(OR)(OR′)(OR′′). Given
that the selectivity in the synthesis of 4 and 6 was
limited, it could be anticipated that problems would arise

Tri- and Tetraalkoxysilanes with Four Different Substituents
J . Org. Chem., Vol. 62, No. 13, 1997 4459
Sch em e 5
F igu r e 2. HPLC of 7 in 5% EtOAc in hexane on Nucleosil
50-5.
Sch em e 4
Sch em e 6
using the same method here. This was also the case.
Sequential substitution to 1, in the presence of pyridine,
(-)-menthol, 2, and phenethyl alcohol as well as many
other alcohol combinations led to rather complex mix-
tures of trialkoxysilanes (Scheme 4). Probably the dif-
ference in reactivity between the two first chlorine atoms
was too small to ensure clean formation of a monoalkoxy-
silane. We therefore turned to a method used previously
by us for synthesis of unsymmetrical dimethyldialkoxy-
silanes.7 The first alcohol, menthol, was reacted with
excess 1, and the excess 1 was removed by evaporation.
Thus it was possible to get a solution of mono(menthyl-
oxy)dichlorosilane in toluene and pyridine. Reaction of
this compound with 2 followed by phenethyl alcohol now
gave the mixed trisiloxane 7 relatively cleanly, and it
could be isolated pure by flash chromatography in 51%
yield.
Compound 7 was a mixture of two diastereoisomers
at silicon. These could not be separated by flash chro-
matography or TLC, but the diastereomeric ratio could
be measured by 13C NMR. At 75 MHz small splittings
of the signals (0.1 ppm) could be identified, consistant
with a diasteomeric ratio of 5:1. This was confirmed by
HPLC (Figure 2). It was interesting to note that 7 was
not only a stable compound and its configuration at
silicon remained unchanged on storage.
with a diastereomeric ratio of 1:2. Thus the diastereo-
isomer that previously had been minor isomer was now
the major one. This suggested that the diastereomeric
ratio was mainly determined by steric bias when ROS-
iMeCl2 reacted with another alcohol. Interestingly this
also suggests that there may be cases where alkoxysi-
lanes with opposite stereochemistry at silicon can be
made by a simple change in the order of addition of the
reagents (in a one-pot reaction).
Com p ou n d s of Typ e Si(OR)3(OR′). Stepwise reac-
tion of 3 molar equiv of 2 and 1 molar equiv of phenethyl
alcohol with tetrachlorosilane (8) in the presence of
pyridine gave the expected tetraalkoxysilane 9 in 82%
yield (Scheme 5). Similarly reaction of 8 with 3 molar
equiv of menthol and 1 molar equiv of 2-methoxybenzyl
alcohol gave the tetraalkoxysilane 10 in 64% yield
(Scheme 6). This type of unsymmetrically substituted
tetraalkoxysilane could potentially be useful as a chiral
protection group. This was further explored.
We found that the tris(menthyloxy)silyl chloride 11
was quite stable and could be isolated. From 8 and 3
molar equiv of (-)-menthol, the trialkoxysilyl chloride 11
could be obtained in 93% yield (Scheme 7). This com-
pound proved to be stable toward chromatography.
(Interestingly this stability was very structure dependent
as the corresponding trialkoxysilyl chlorides made from
2 or 5 decomposed during chromatography.) Compound
11 could be used as a reagent to silylate alcohols.
Compound 11 reacted with 2-methoxybenzyl alcohol in
pyridine and toluene to give 10 in 96% yield. Protection
of the racemic alcohol 129 with 11 gave the silylated
Nucleophilic substitution of silicon halides with oxygen
nucleophiles generally gives inversion.8 Thus the dia-
stereomeric ratio in 7 should in principle depend on the
order of addition of the alcohols. This was found to be
true. When 1 was reacted with 2 first and then with (-)-
menthol and finaly phenethyl alcohol, 7 was obtained
(8) Sommer, L. H. Stereochemistry, Mechanism and Silicon; McGraw-
Hill Inc.: New York, 1965.

4460 J . Org. Chem., Vol. 62, No. 13, 1997
Clausen and Bols
Sch em e 7
Sch em e 9
Sch em e 8
Sch em e 10
product 13 in 84% yield as a 1:1 mixture of diastereo-
isomers (Scheme 8). Note that the chiral silyl chloride
had little selectivity for either enantiomer of 12. The
diastereomers could not be separated on TLC or coloumn,
but were distinguishable in the 13C NMR spectrum.
Surprisingly there were no differences in the shifts from
the pyridazine ring, but a 0.1 ppm splitting was observed
at the C-1, C-2, and C-5 (the chiral centers) of the
menthyloxy groups. Thus 11 could be used as a simple
reagent for determining the ee of an alcohol.
Sch em e 11
The effect of using trialkoxysilyl groups as chiral
auxillaries in a reaction was briefly explored. We carried
out Diels-Alder reactions with two silylated sorbyl
alcohols. Reaction of 11 with sorbyl alcohol gave chiral
diene 14 in 85% yield (Scheme 9), while reaction of 8 with
first 3 mol of 2 and then sorbyl alcohol gave 17 in 64%
yield (Scheme 10). Diels-Alder reaction of 14 and 17
with 4-phenyl-1,2,4-triazoline-3,5-dione and maleic an-
hydride was then carried out (Scheme 11). The cycload-
ducts 15, 16, 18, and 19 were obtained in good to
excellent yield, but with no or in one case (18) low
stereoselectivity (Scheme 11). The diastereomeric (pseudo-
enantiomeric) ratio was readily determined by 13C NMR.
The cycloadducts of maleic anhydride (16 and 19) were
obtained with total diastereomeric control with respect
to giving the expected endo adducts.
The 13C NMR spectra of 17 and 18 showed differences
in the signals of the two diastereoisomers of up to 0.4
ppm, while in 13, 15, and 16 only up to 0.1 ppm
differences between diastereomeric signals was observed.
This suggested that the fructose moity influenced its
surroundings much more profoundly than a menthyloxy
group. This could be the cause of the better stereoselec-
tivity obtained with this group.
(9) Bols, M.; Hazell, R.; Thomsen, I. Chem. Eur. J . 1997, 3, 940-7.

Tri- and Tetraalkoxysilanes with Four Different Substituents
J . Org. Chem., Vol. 62, No. 13, 1997 4461
Sch em e 12
Sch em e 14
Sch em e 13
in reasonable yield. The reaction with several alcohols
gave complex mixtures suggesting that the difference in
reactivity of the first, second, and third chlorine atom
was quite small. We therefore employed the sterically
demanding methyl 2,4,6-tri-O-benzyl-R-D-glucopyrano-
side as the first alcohol, anticipating that this group
would slow further substitution.
This compound was reacted with excess 8 and pyridine,
and the excess of 8 was removed by distillation. Then 1
equiv of 2, 2-methoxybenzyl alcohol, and finaly phenethyl
alcohol were added. This gave tetraalkoxysilane 23 in
62% yield after flash chromatography (Scheme 14). This
compound had a chiral silicon atom, and 13C NMR
revealed that two diastereoisomers at silicon were present
in a ratio of 5:1, which was confirmed by HPLC. The
configuration remained constant on storage.
It is remarkable that an alkoxysilane of this complexity
could be made in a one-pot reaction in so good a yield.
This emphasizes the high degree of selectivity that can
be obtained and the great possibilities of engineering
chiral silyl groups from alkoxysilanes. Thus it may, in
principle, be possible to synthesize chiral auxillaries from
one set of chiral alcohols with opposite stereochemistry
at silicon simply by a change in addition of reactants.
Such pseudo-enantiomeric compounds might very well
induce opposite enantioselectivity in a reaction, and this
would be achieved without requiring the antipodes of the
chiral alcohols.
All of the compounds prepared above were relatively
insensitive to hydrolysis. They did however slowly
hydrolyze in 10% aqueous THF. In the case of the
comparable compounds 3 and 9, hydrolysis of the tet-
raalkoxysilane 9 was faster. Hydrolysis of 9 was com-
plete in 5 d at 25 °C, while conversion of 3 was 50% in
the same timespan. This also suggests that oligoalkoxy-
silyl groups have a stability comparable to that of
trimethylsilyl groups.
Con clu sion . This work has surprisingly shown that
unsymmetrical alkoxysilanes are stable compounds that
can be isolated in pure form and do not undergo spon-
taneous decomposition or disproportionation. Even tri-
and tetraalkoxysilanes with four different substituents
could be made. They could readily be prepared from
oligochlorosilanes by sequential addition of alcohols in
the presence of pyridine. The fact that this type of
compound is so readily available opens tremendous
opportunities of using them as chiral protection groups
or chiral tethers. In both cases it can be anticipated that
It may seem surprising that some stereoselectivity was
observed in the reaction of 17 with 4-phenyl-1,2,4-
triazoline-3,5-dione and not in the reaction of 17 with
maleic anhydride, since the latter reaction was much
slower which should favor selectivity. Inspection of
molecular models of 18 and 19 suggested, however, that
steric interaction between the Ph group and the fructose
derivatives would occur in formation of 18. The poor
stereoselectivity (pseudo-enantioselectivity) in these reac-
tions is probably a result of the freedom of rotation the
silyl group and its substituents have, as different rota-
mers may induce different possibly opposite selectivity.
More rigid silicon substituents (such as diols) might
improve this.
Com p ou n d s of Typ e Si(OR)2(OR′)2. An example of
this type of alkoxysilane was prepared by reacting 8 with
2 mol of readily available methyl 2,4,6-tri-O-benzyl-R-D-
glucopyranoside followed by benzyl alcohol. Tetraalkoxy-
silane 20 was obtained in 60% yield (Scheme 12). Note
that steric crowding around silicon is quite severe in 20.
Com p ou n d s of Typ e Si(OR)2(OR′)(OR′′). Two com-
pounds of this type were made. First 8 was reacted with
2 mol of menthol, then 2-propanol and finaly 2-methoxy-
benzyl alcohol to give tetraalkoxysilane 21 in 59% yield.
When 2-propanol was substituted with 2-methoxybenzyl
alcohol, and 2-methoxybenzyl alcohol was substituted
with allyl alcohol, 53% of 22 was obtained (Scheme 13).
The control of synthesis and stability of this type of
derivative was particularly noteworthy as it allows
precursors for tethered reactions to be made where R′OH
and R′′OH were reactants. This opens opportunities for
a very easy way of carrying out enantioselective tethered
reactions.
Com p ou n d s of Typ e Si(OR)(OR′)(OR′′)(OR*). This
was the most difficult tetraalkoxysilane to obtain pure

4462 J . Org. Chem., Vol. 62, No. 13, 1997
Clausen and Bols
pyridine. The solution was left stirring for 2 h whereupon 0.15
g (2.59 mmol) of dry allyl alcohol was added, and the reaction
was left stirring overnight. The mixture was filtered and
concentrated. Flash column chromatography of the residue
in Et2O/petroleum ether (1:4) gave in the first fraction 0.65 g
of 6 (41%) as viscous liquid. A second fraction of 0.39 g of tris-
(1,2:5,6-diisopropylideneglucofuranos-3-oxy)methylsilane (6a ,
19%) was also obtained. 6. 1H NMR (CDCl3): δ 5.9 (m, 1H),
5.84 (2d, 2H), 5.23 (dd, 1H, J ) 17 Hz), 5.07 (dd, 1H, J ) 11
Hz), 4.57 (bs, 2H), 4.46 (d, 1H, J ) 3.3 Hz), 4.42 (d, 1H, J )
3.8 Hz), 4.30 (m, 2H), 4.23 (m, 2H), 4.1-3.9 (m, 6H), 1.44 (s,
6H), 1.36 (s, 6H), 1.29 (s, 6H), 1.26 (d, 6H), 0.22 (s, 3H). 13C
NMR (CDCl3): δ 136.0, 114.8, 111.7, 111.6, 109.0, 108.9, 105.1,
85.2, 81.8, 81.6, 75.2, 75.1, 72.2, 67.4, 67.3, 63.6, 26.7, 26.2,
25.1, 25.0, -6.7. MS (EI) m/z 618 (M+). 6a . 1H NMR
(CDCl3): δ 5.87 (d, 3H, J ) 3.9 Hz), 4.61 (d, 3H, J ) 2.2 Hz),
4.47 (d, 3H, J ) 3.9 Hz), 4.24 (m, 3H), 4.1-3.96 (m, 9H), 1.46
(s, 9H), 1.38 (s, 9H), 1.31 (s, 9H), 1.27 (s, 9H), 0.29 (s, 3H).
Met h yl(((1R,2S,5R)-2-isop r op yl-5-m et h ylcycloh exyl)-
oxy)((1,2:4,5-d iisop r op ylid en efr u ctop yr a n os-3-yl)oxy)(2-
p h en yleth oxy)sila n e (7). To 1 mL of H3CSiCl3 (1, 8.5 mmol)
and 0.20 g of (-)-menthol (1.28 mmol) in 25 mL of dry toluene
was added 0.2 mL of dry pyridine, and the mixture was stirred
for 1 h. Half the volume of the solution was removed by
distillation and then refilled with toluene. Then 0.35 g of 1,2:
4,5-diisopropylidenefructopyranose (2, 1.35 mmol) and 0.2 mL
of dry pyridine were added at 0 °C, and the mixture was left
overnight at rt. Next 0.20 mL of phenethyl alcohol (1.67 mmol)
and 0.2 mL of pyridine was added, and the mixture was again
stirred for 2 d. Filtration and evaporation followed by column
chromatography (Et2O/petroleum ether (1:4)) gave 0.39 g of 7
(51%). HPLC (Nucleosil 50-5, EtOAc-hexane 1:19) and 13C
NMR showed a 1:5 mixture of diastereomers. Major stereo-
isomer. 1H NMR (CDCl3): δ 7.31-7.20 (m, 5H), 4.23-3.86
(m, 9H), 3.71 (dt, 1H, J ) 4.4 Hz), 2.87 (t, 2H, J ) 7.5 Hz), 2.2
(m, 1H), 1.96 (bd, 1H), 1.62 (dt, 2H), 1.51, 1.49, 1.42, 1.35 (4s,
12H), 1.1-0.8 (m, 14H), 0.19 (s, 3H). 13C NMR (CDCl3): δ
138.8, 129.0, 128.2, 126.1, 111.8, 108.8, 105.0, 77.7, 73.8, 72.5,
71.6, 70.9, 63.5, 60.2, 49.7, 44.9, 39.0, 34.4, 31.6, 28.1, 26.6,
26.3, 26.2, 25.4, 22.8, 22.2, 21.1, 15.7, -5.8. Minor stereoiso-
mer. 13C NMR (CDCl3): identical to major except for δ 77.6,
73.7, 72.4, 71.0, 45.0, 25.2, 22.7, 15.6. MS (PDMS) m/z 578
(M+), 563 (M - CH3).
Alternatively, to 1 mL of H3CSiCl3 (1, 8.5 mmol) and 0.331
g of 2 (1.27 mmol) in 20 mL of dry toluene was added 0.2 mL
of dry pyridine, and the mixture was stirred for 2 h. Half the
volume of the solution was removed by distillation and then
refilled with toluene. Then 0.2 g of (-)-menthol (1.28 mmol)
and 0.2 mL of dry pyridine was added at 0 °C, and the mixture
was left overnight at rt. Next 0.20 mL of phenethyl alcohol
(1.67 mmol) and 0.2 mL of pyridine were added, and the
mixture was stirred for 18 h. Filtration and evaporation
followed by column chromatography (Et2O/petroleum ether (1:
4)) gave 0.33 g of 7 (43%), but in this case HPLC (Nucleosil
50-5, EtOAc-hexane 1:19) and 13C NMR showed a 2:1 mixture
of diastereomers.
(2-P h en yleth oxy)tr is((1,2:4,5-d iisop r op ylid en efr u cto-
p yr a n os-3-yl-oxy)sila n e (9). To 0.68 g of 1,2:4,5-diisopro-
pylidenefructopyranose (2, 2.62 mmol) and 0.10 mL of SiCl4
(0.87 mmol) in 25 mL of toluene was added 0.4 mL of pyridine,
and the solution was stirred 1 h at 50 °C. Then 0.106 g of
phenethyl alcohol (0.87 mmol)) was added at rt, and the
mixture was stirred overnight. Filtration and column chro-
matography (Et2O/petroleum ether (1:4)) gave crystalline 9 in
82% yield (0.76 g). The material could be recrystallized from
MeCN. Mp 150 °C. 1H NMR (CDCl3): δ 7.3-7.1 (m, 5H), 4.3-
4.1 (m, 17H), 3.96 (d, 3H, J 1a,1b ) 13 Hz), 3.90 (d, 3H, J 5,6 ) 7
Hz), 2.93 (t, 2H), 1.52 (s, 3H), 1.47 (s, 3H), 1.40 (s, 3H), 1.32
(s, 3H). MS (EI) m/z 926 (M+).
considerable investigation is required to identify those
alkoxy groups necessary to obtain good stereoselectivity.
We have demonstrated the use of chiral trialkoxysilyl
groups as chiral protection groups, but in those cases no
or small stereoinduction was obtained. The chiral aux-
ilaries employed are perhaps too flexible, and the use of
more rigid systems (such as, for example, C-2 symmetric
diols as alkoxy substituents) might very well improve
this. Research is in progress in our laboratory to try to
solve this problem as well as investigations of the
performance of these compounds in many other applica-
tions.
However one interesting observation was that the
chiral trialkoxysilyl groups could be used for simple
determination of the enantiomeric ratio of an alcohol by
13C NMR. It was found that when the trialkoxysilyl
group contained a carbohydrate derivative, particularly
2, the largest differences between the 13C signals of
diastereoisomers were observed. With fructose groups
differences of up to 0.5 ppm were observed on individual
signals. As these compounds are extremely easy to
prepare and have a stability close to a trimethylsilyl
(TMS) group, they may find use as a chiral equivalent of
a TMS group.
Exp er im en ta l Section
General experimental procedures were as previously de-
scribed.9 The pyridine was dried over KOH.
Met h ylt r is(1,2:4,5-d iisop r op ylid en efr u ct op yr a n os-3-
yl)oxy)sila n e (3). To 0.100 mL of H3CSiCl3 (1, 0.85 mmol)
and 0.663 g of 1,2:4,5-diisopropylidenefructopyranose (2, 2.55
mmol) in 25 mL of dry toluene was added 0.25 mL of dry
pyridine, and the solution was stirred overnight. The solution
was washed with water (5 mL), dried (MgSO4), and evapo-
rated. Chromatography in Et2O/petroleum ether (1:4) gave
0.483 g of 3 (70%). The material could be recrystallized in
MeOH. Mp 173 °C. 1H NMR (CDCl3): δ 4.33 (d, 3H, J ) 8.8
Hz), 4.08-4.17 (m, 9H), 3.98 (d, 3H, J ) 13.2 Hz), 3.90 (d, 3H,
J ) 7.2 Hz), 3.88 (d, 3H), 1.53 (s, 3H), 1.46 (s, 3H), 1.37 (s,
3H), 1.32 (s, 3H), 0.49 (s, 3H). 13C NMR (CDCl3): δ 111.8 (3C),
108.9 (3C), 104.8 (3C), 77.6 (3C), 73.8 (3C), 71.5 (3C), 71.2 (3C),
60.2 (3C), 28.2 (3C), 26.6 (3C), 26.2 (3C), 26.1 (3C), -4.8. MS
(EI) m/z 820 (M+).
Bis(1,2:4,5-d iisop r op ylid en efr u ct op yr a n os-3-yl)oxy)-
isop r op oxym eth ylsila n e (4). To 0.59 mL of H3CSiCl3 (1, 5.0
mmol) and 2.6 g of 1,2:4,5-diisopropylidenefructopyranose (10.0
mmol) in 25 mL of dry toluene was added 2 mL of dry pyridine,
and the solution was stirred overnight. Then 0.38 mL (0.3 g,
5.0 mmol) of dry 2-propanol was added, and the mixture was
left overnight. The precipitated pyridine hydrochloride was
removed by filtration and the solution evaporated. Purification
by flash column chromatography in Et2O/petroleum ether (1:
4) gave in the second fraction 2.19 g of 4 (70%) as a viscous
liquid. The first fraction was 141 mg of diisopropoxy((1,2:4,5-
diisopropylidenefructopyranos-3-yl)oxy)methylsilane (4a , 7%).
A third fraction of 303 mg of 3 (7%) was also isolated. 4. 1H
NMR (CDCl3): δ 4.22-4.15 (m, 3H), 4.1-4.0 (m, 6H), 3.95-
3.78 (m, 6H), 1.49, 1.43, 1.40, 1.39 1.33, 1.29, 1.25, 1.23 (8s,
24H), 1.1 (d, 6H), 0.26 (s, 3H). 13C NMR (CDCl3): δ 111.7,
111.5, 108.7, 108.6, 104.8, 104.7, 77.5, 77.4, 73.7, 73.6, 71.6,
71.3, 71.1, 70.6, 65.2, 60.2, 60.0, 28.1, 28.0, 26.6, 26.3, 26.2,
26.1, 26.0, 25.9, 25.4, -5.0. MS (EI) m/z 620 (M+). 4a . 1H
NMR (CDCl3): δ 4.25-4.08 (m, 6H), 3.97 (d, 1H, J 1a,1b ) 13
Hz), 3.89 (d, 1H, J 5,6 ) 6 Hz), 3.87 (d, 1H, J 3,4 ) 8 Hz), 1.49,
1.46, 1.38, 1.32 (4s, 12H), 1.15 (2d, 12H), 0.18 (s, 3H). 13C NMR
(CDCl3): δ 111.8, 108.7, 105.0, 77.6, 73.8, 71.7, 70.8, 64.9, 60.2,
28.2, 26.6, 26.3, 26.1, 25.5, 25.4 (2s, 4C), -5.1.
Ch lor otr is(((1R,2S,5R)-2-isopr opyl-5-m eth ylcycloh exyl)-
oxy)sila n e (11). To 2 mL of SiCl4 (8, 17.4 mmol) and 8.17 g
of (-)-menthol (52 mmol) in 80 mL of toluene was added 4.5
mL of pyridine, and the mixture was stirred 4 h at 50 °C. Then
100 mL of petroleum ether was added, and the mixture was
filtered through a short column of silica and eluted with
petroleum ether. Evaporation of the eluate gave 6.5 g of trimer
(Allyloxy)bis((1,2:5,6-d iisop r op ylid en eglu cofu r a n os-3-
yl)oxy)m eth ylsila n e (6). To 0.30 mL of H3CSiCl3 (1, 2.54
mmol) and 1.33 g of 1,2:5,6-diisopropylideneglucofuranose (5,
5.11 mmol) in 50 mL of dry toluene was added 0.5 mL of dry

Tri- and Tetraalkoxysilanes with Four Different Substituents
J . Org. Chem., Vol. 62, No. 13, 1997 4463
11 (70%) as an oil, which was sufficiently pure for further use.
1H NMR (CDCl3): δ 3.75 (dt, 3H), 2.2 (dp, 3H), 2.1 (bd, 3H),
1.6 (m, 6H), 1.4 (m, 3H), 1.1-0.8 (m, 12H), 0.9 (2d, 18H), 0.8
(d, 9H). 13C NMR (CDCl3): δ 74.5, 49.5, 44.3, 34.4, 31.6, 25.3,
22.7, 22.1, 21.2, 15.7. MS (EI) m/z 528 (M+), 530 (M + 2).
((2-Meth oxyben zyl)oxy)tr is(((1R,2S,5R)-2-isop r op yl-5-
m eth ylcycloh exyl)oxy)sila n e (10). To 0.10 mL of SiCl4
(0.87 mmol) and 0.41 g of (-)-menthol (2.62 mmol) in 25 mL
of toluene was added 0.25 mL of pyridine, and the mixture
was stirred for 1 h at 50 °C. Then 0.120 g of 2-methoxybenzyl
alcohol (0.87 mmol) was added, and the reaction was left
overnight, while cooling to rt. Filtration and evaporation
followed by column chromatography in CH2Cl2/petroleum ether
(1:4) gave 354 mg (64%) of 10. 25 mg of bis((2-methoxybenzyl)-
oxy)bis(((1R,2S,5R)-2-isopropyl-5-methylcyclohexyl)oxy)-
silane was obtained as the second fraction. 10. 1H NMR
(CDCl3): δ 7.5 (d, 1H), 7.2 (t, 1H), 7.0 (t, 1H), 6.8 (d, 1H), 4.9
(s, 2H), 3.8 (s, 3H), 3.7 (dt, 3H), 2.3 (dp, 3H), 2.1 (bd, 3H), 1.6
(m, 6H), 1.4 (m, 3H), 1.2-0.8 (m, 12H), 0.9 (2d, 18H), 0.8 (d,
9H). MS (EI) m/z 630 (M+).
Alternatively 0.276 g of (2 mmol) 2-methoxybenzyl alcohol
and 2.2 g of 11 (4.15 mmol) in 40 mL of toluene was added to
0.4 mL of pyridine, and the mixture was stirred for 3 h.
Filtration and evaporation followed by column chromatography
in CH2Cl2/petroleum ether (1:4) gave 1.21 g (96%) of 10.
(((8-P h en yl-7,9-d ioxo-1,6,8-t r ia za [4.3.0]b icyclon on -3-
en -2-yl)m eth yl)oxy)tr is(((1R,2S,5R)-2-isop r op yl-5-m eth -
ylcycloh exyl)oxy)sila n e (13). To a solution of 0.260 g of 127
(1.0 mmol) and 1.0 g of 11 (1.89 mmol) in 20 mL of toluene
was added 0.2 mL of pyridine, and the solution was stirred
overnight. The solution was filtered through silica with
toluene, evaporated, and column chromatographed in Et2O/
petroleum ether (1:2) giving 0.630 g of 13 (84%). 1H NMR
(CDCl3): δ 7.5-7.2 (m, 5H), 5.95 (d, 2H), 4.55 (bs, 1H), 4.2 (d,
1H), 4.0 (m, 3H), 3.8 (m, 3H), 2.1 (m, 3H), 1.9 (bd, 3H), 1.5
(dt, 6H), 1.2-0.8 (m, 15H), 0.8 (2d, 18H), 0.7 (d, 9H). 13C NMR
(CDCl3): δ 152, 150, 131, 128.8, 127.7, 125.1, 123.8, 120, 73.4,
73.3 (3C), 62.1, 54.7, 49.6, 49.5 (3C), 44.7 (3C), 43.3, 34.4 (3C),
31.6, 31.5 (3C), 25.2 (3C), 22.6 (3C), 22.2 (3C), 21.1 (3C), 15.5
(3C). MS (PDMS) m/z 751 (M+).
((E,E)-Hexa -2,4-d ien yloxy)tr is(((1R,2S,5R)-2-isop r op yl-
5-m eth ylcycloh exyl)oxy)sila n e (14). To 1.5 g of 11 (2.83
mmol) and 0.20 mL of sorbyl alcohol (1.77 mmol) in 20 mL of
toluene was added 3 mL of pyridine, and the mixture was left
stirring overnight. The solution was filtered through silica
with toluene, and the filtrate was evaporated. The residue
was purified using column chromatography in petroleum ether.
First eluted was the starting material when eluting with
petroleum ether, whereafter 14 was eluted with 1% Et2O in
petroleum ether. Yield of 14: 0.890 g (85%). 1H NMR
(CDCl3): δ 6.3-6.0 (m, 2H), 5.75-5.55 (m, 2H), 4.35 (d, 2H),
3.7 (dt, 3H), 2.3 (dp, 3H), 2.1 (bd, 3H), 1.8 (d, 3H), 1.6 (m, 6H),
1.4-1.0 (m, 15H), 0.9 (2d, 18H), 0.8 (d, 9H). 13C NMR
(CDCl3): δ 131.2, 130.3, 129.3, 128.6, 73.3 (3C), 63.5, 49.8 (3C),
44.8 (3C), 34.5 (3C), 31.7 (3C), 25.2 (3C), 22.7 (3C), 22.2 (3C),
21.2 (3C), 18.1, 15.6 (3C).
((((2,5-cis)-5-Meth yl-8-ph en yl-7,9-dioxo-1,6,8-tr iaza[4.3.0]-
b icyclon on -3-en -2-yl)m et h yl)oxy)t r is(((1R,2S,5R)-2-iso-
p r op yl-5-m eth ylcycloh exyl)oxy)sila n e (15). A solution of
30 mg of N-phenylurazole (0.17 mmol) and 18 mg of t-BuOCl
(0.17 mmol) in 2 mL of EtOAc was degassed by being purged
with N2 for 15 min. Then 100 mg of 14 (0.17 mmol) dissolved
in 5 mL of EtOAc was added to the red solution, and the red
color disappeared. The solution was evaporated, and the
residue was column chromatographed in Et2O/petroleum ether
(1:4) giving 121 mg (93%) of 15. 1H NMR (CDCl3): δ 7.4 (m,
5H), 6.0 (d, 1H), 5.9 (dd, 1H), 4.5 (bs, 2H), 4.35 (2dd, 1H), 4.05
(2dd, 1H), 3.7 (dt, 3H), 2.2 (dp, 3H), 2.0 (bd, 3H), 1.6 (d, 6H),
1.5 (d, 3H), 1.4-1.0 (m, 15H), 0.9 (2d, 18H), 0.8 (d, 9H). 13C
NMR (CDCl3): δ 151.6, 128.9, 127.8, 127.6, 127.5, 125.3, 123.0,
73.4, 73.3 (3C), 63.2, 55.5, 55.4, 50.2, 50.1, 49.6 (3C), 44.7, 44.6
(3C), 34.4 (3C), 31.6 (3C), 25.2 (3C), 22.6 (3C), 22.2 (3C), 21.2
(3C), 19.1, 15.6 (3C). MS (PDMS) m/z 765 (M+).
of CH2Cl2 was added 17 mg of maleic anhydride (0.17 mmol),
and the solution was left stirring in the dark for 7 d. The
solvent was evaporated, and the residue was column chro-
matographed in Et2O/petroleum ether (1:4) giving 98 mg (84%)
of 16. 1H NMR (CDCl3): δ 5.95 (m, 1H), 5.8 (dt, 1H), 4.25
(dd, 1H, J 6′a6′b ) 10.4, J 66′a ) 7.1 Hz), 4.15 (2dd, 1H), 3.7 (dt,
3H), 3.5 (dd, 1H, J 1,5 ) 9.3, J 5,6 ) 6 Hz), 3.3 (dd, 1H, J 1,9 ) 7.4
Hz) 2.5 (m, 2H), 2.3 (p, 3H), 2.0 (bd, 3H), 1.6 (d, 6H), 1.4 (d,
3H), 1.4-1.0 (m, 15H), 0.9 (2d, 18H), 0.8 (d, 9H). 13C NMR
(CDCl3): δ 134.4, 134.3, 130.9, 73.4 (3C), 62.5, 49.6 (3C), 45.9,
44.7 (3C), 43.0, 42.9, 38.5, 38.4, 34.4 (3C), 31.6 (3C), 30.7, 30.6,
25.2 (3C), 22.6 (3C), 22.2 (3C), 21.2 (3C), 16.4, 15.6 (3C). MS
(EI) m/z 688 (M+).
((E,E)-Hexa-2,4-dien yloxy)tr is((1,2:4,5-diisopr opyliden e-
fr u ctop yr a n os-3-yl)oxy)sila n e (17). To 1.36 g of 1,2:4,5-
diisopropylidenefructopyranose (2, 5.23 mmol) and 0.2 mL of
SiCl4 (8, 1.74 mmol) was added 0.8 mL of pyridine, and the
mixture was stirred for 3 h at 50 °C. Then 0.27 mL of sorbyl
alcohol (2.38 mmol) was added. The solution was filtered and
evaporated. Column chromatography in Et2O/petroleum ether
(1:3) gave first bis((E,E)-hexa-2,4-dienyloxy)bis(1,2:4,5-diiso-
propylidenefructopyranos-3-oxy)silane (230 mg, 18%) and
second 1.00 g (64%) of 17. 1H NMR (CDCl3): δ 6.1 (m, 2H),
5.6 (m, 2H), 4.4 (d, 2H), 4.2-3.9 (m, 21H), 1.7 (d, 3H), 1.5 (s,
9H), 1.45 (s, 9H), 1.35 (s, 9H), 1.3 (s, 9H). 13C NMR (CDCl3):
δ 131.0, 130.9, 129.2, 128.9, 111.4 (3C), 108.9 (3C), 104.5 (3C),
76.6 (3C), 73.6 (3C), 72.0 (3C), 71.9 (3C), 64.4, 60.8 (3C), 27.9
(3C), 26.3 (6C), 26.0 (3C), 18.0. MS (EI) m/z 902 (M+).
((((2,5-cis)-5-Meth yl-8-ph en yl-7,9-dioxo-1,6,8-tr iaza[4.3.0]-
b icyclon on -3-en -2-yl)m et h yl)oxy)t r is((1,2:4,5-d iisop r o-
p ylid en efr u ct op yr a n os-3-yl)oxy)sila n e (18). N-phenyl-
urazol (30 mg) and 25 mg of tert-butyl hypochlorite in 10 mL
of EtOAc was degassed by being purged with N2 for 15 min.
Then 150 mg of 17 dissolved in EtOAc was added to the red
solution causing the red color to disappear. The solution was
evaporated, and the residue was column chromatographed in
Et2O/petroleum ether (3:1) giving 175 mg (97%) of 18. 1H
NMR (CDCl3): δ 7.4 (m, 5H), 6.1 (m, 1H), 5.8 (m, 1H), 4.6 (m,
2H), 4.5 (m, 2H), 4.3-3.9 (m, 7H), 1.55 (d, 9H), 1.5 (d, 3H),
1.4 (s, 3H), 1.35 (s, 9H), 1.25 (s, 9H). 13C NMR (CDCl3): major
isomer, δ 152.0, 151.5, 131.2, 128.9, 127.8, 127.6, 125.3, 123.0,
111.3 (3C), 108.9 (3C), 104.4 (3C), 76.4 (3C), 73.6 (3C), 72.3
(3C), 72.1 (3C), 63.9, 60.8 (3C), 55.2, 50.6, 27.9 (3C), 26.4 (3C),
26.1 (3C), 26.0 (3C), 19.1; minor isomer, δ 152.4, 151.1, 131.2,
128.9, 127.8, 127.6, 125.5, 122.8, 111.4 (3C), 108.8 (3C), 104.5
(3C), 76.4 (3C), 73.7(3C), 72.4(3C), 72.0 (3C), 64.0, 60.7 (3C),
54.8, 51.0, 28.0 (3C), 26.4 (3C), 26.1 (3C), 26.0 (3C), 19.4. MS
(EI) m/z 1077 (M+).
(((5-Met h yl-7,9-d ioxo-8-oxa [4.3.0]b icyclon on -3-en e-2-
yl)m eth yl)oxy)tr is(1,2:4,5-diisopr opyliden efr u ctopyr an os-
3-yl)oxy)sila n e (19). To 150 mg of 17 (0.17 mmol) in 2 mL
of EtOAc was added 16 mg of maleic anhydride (0.16 mmol),
and the mixture was left to stir in the dark for 7 d. The solvent
was evaporated, and the residue was subjected to column
chromatography (Et2O/petroleum ether (1:1)) to give 100 mg
(60%) of 19. 13C NMR (CDCl3): δ 171.4, 134.3, 134.2, 131.3,
131.0, 111.53, 111.48 (3C), 108.90, 108.87 (3C), 104.60, 104.56
(3C), 76.7 (3C), 73.6 (3C), 72.4 (3C), 72.1, 72.0 (3C), 63.6, 63.4,
60.84, 60.78 (3C), 46.0, 45.9, 43.0, 42.8, 38.2, 38.1, 30.6, 30.5,
28.0 (3C), 26.4 (3C), 26.1 (3C), 26.0 (3C), 16.4. MS (EI) m/z
1024 (M + Na).
Bis(ben zyloxy)bis((1-O-m eth yl-2,4,6-tr i-O-ben zyl-r-D-
glu cop yr a n os-3-yl)oxy)sila n e (20). To a solution of 125 mg
of methyl 2,4,6-tri-O-benzyl-R-D-glucopyranoside in 10 mL of
toluene was added first 0.17 mL of a 9.1% SiCl4 (v/v) in toluene
solution and then 0.2 mL of pyridine. After the solution was
stirred for 1 h, 0.03 mL of BnOH was added to the reaction
mixture, and the mixture was left to stir overnight. The
mixture was filtered and concentrated. Column chromatog-
raphy in Et2O/petroleum ether (1:1) gave 95 mg (60%) of 20.
1H NMR (CDCl3): δ 7.2-7.4 (m, 40H), 5.1 (d, 2H), 4.9 (d, 4H),
4.4-4.8 (m, 14H), 3.4-3.8 (m, 10H), 3.3 (s, 6H). 13C NMR
(CDCl3): δ 140.4, 138.7, 137.8, 128.3, 128.2, 128.0, 127.8, 127.6,
127.5, 127.4, 127.1, 126.5, 126.3, 98.0, 80.0, 78.4, 75.4, 74.2,
73.4, 73.1, 69.6, 68.6, 65.1, 54.9.
(((5-Meth yl-7,9-d ioxo-8-oxa [4.3.0]bicyclon on -3-en -2-yl)-
m eth yl)oxy)tr is(((1R,2S,5R)-2-isop r op yl-5-m eth ylcyclo-
h exyl)oxy)sila n e (16). To 100 mg of 14 (0.17 mmol) in 2 mL

4464 J . Org. Chem., Vol. 62, No. 13, 1997
Clausen and Bols
glu cop yr a n os-3-yl)oxy)(2-p h en yleth oxy)sila n e (23).
Bis(((1R,2S,5R)-2-Isop r op yl-5-m eth ylcycloh exyl)oxy)-
((2-m eth oxyben zyl)oxy)isop r op oxysila n e (21). To a solu-
tion of 0.10 mL of SiCl4 (8, 0.87 mmol) and 272 mg of (-)-
menthol (1.74 mmol) in 15 mL of toluene was added 0.2 mL of
pyridine at -78 °C, and the solution was left stirring for 15
min, whereafter it was allowed to warm to rt. Then 52 mg of
2-propanol (0.87 mmol) and 0.1 mL of pyridine in 4 mL toluene
were added. The mixture was stirred for 20 min, and then
120 mg of 2-methoxybenzyl alcohol (0.86 mmol) and 0.1 mL
of pyridine were added. The mixture was stirred 18 h and
then filtered and concentrated. Column chromatography in
ether-pentane (1:1) gave 275 mg (59%) of 21. 13C NMR
(CDCl3): δ 156.1, 129.3, 127.5, 127.1, 120.2, 109.5, 73.28, 73.25
(2C), 65.9, 60.1, 55.0, 49.7 (2C), 44.7 (2C), 34.5 (2C), 31.6 (2C),
25.3 (2C), 25.2 (2C), 22.8 (2C), 22.2 (2C), 21.2 (2C), 15.7 (2C).
MS (EI) m/z 534 (M+).
(Allyloxy)b is(((1R,2S,5R)-2-isop r op yl-5-m et h ylcyclo-
h exyl)oxy)((2-m eth oxyben zyl)oxy)sila n e (22). A solution
of 0.10 mL of SiCl4 (8, 0.87 mmol) in 15 mL of toluene was
slowly added to a solution of 272 mg of (-)-menthol (1.74
mmol) and 0.15 mL of pyridine in 5 mL of toluene. The
solution was left stirring for 1 h, and then 120 mg of
2-methoxybenzyl alcohol (0.86 mmol) and 0.1 mL of pyridine
in 5 mL of toluene were added. The mixture was stirred
overnight, and then a solution of 51 mg of allyl alcohol (0.88
mmol), and 0.1 mL of pyridine in 2 mL of toluene was added.
The mixture was stirred overnight and then filtered and
concentrated. Column chromatography in ether-pentane (1:
1) gave 248 mg (53%) of 22. 1H NMR (CDCl3): δ 7.4 (d, 1H),
7.1 (t, 1H), 6.9 (d, 1H), 6.7 (d, 1H), 5.9 (m, 1H), 5.2 (dd, 1H),
5.0 (dd, 1H), 4.8 (s, 2H), 4.2 (m, 2H), 3.7 (s, 3H), 3.6 (h, 2H),
2.2 (m, 2H), 2.0 (m, 2H), 1.5 (m, 4H), 1.3 (m, 2H), 1.1 (m, 2H),
0.8-1.0 (m, 6H), 0.8 (d, 12H), 0.7 (d, 6H). MS (EI) m/z 532
(M+).
A
solution of 1.0 mL of SiCl4 (8) in 35 mL of toluene was mixed
with 465 mg of methyl 2,4,6-tri-O-benzyl-R-D-glucopyranoside
(1 mmol) and 0.2 mL of pyridine and stirred for 2 d. The
solution was concentrated to half volume by distillation at
reduced pressure, and 260 mg of 2 (1 mmol) and 0.2 mL of
pyridine were added. The mixture was stirred for 4 d. Next
138 mg of 2-methoxybenzyl alcohol (1 mmol) was added, and
stirring was continued for 18 h. Finally 122 mg of phenethyl
alcohol (1 mmol) was added, and the mixture was stirred for
4 h. The mixture was then filtered and concentrated. Column
chromatography in ether-pentane (1:2) gave 630 mg (62%) of
23. HPLC (Nucleosil 50-5 in EtOAc-hexane 1:3) revealed a
5:1 mixture of diastereoisomers at silicon. 1H NMR (CDCl3):
δ 7.6 (d, 1H), 7.2-7.4 (m, 21H), 7.0 (t, 1H), 6.8 (d, 1H), 5.15 (s,
2H), 5.1 (d, 1H), 4.9 (d, 1H), 4.6 (m, 6H), 4.5 (d, 1H) 4.2-4.4
(m, 7H), 3.9 (d, 1H), 3.8 (s, 3H), 3.75 (m, 4H), 3.6 (dd, 1H), 3.3
(s, 3H), 2.9 (t, 2H), 1.6 (s, 3H), 1.55 (s, 3H), 1.5 (s, 3H), 1.4 (m,
3H). 13C NMR (CDCl3): major stereoisomer, δ 155.8, 138.6,
138.5, 138.4, 137.7, 126.9-128.8, 125.8, 120.0, 111.5, 109.1,
108.6, 104.6, 97.7, 79.8, 78.1, 76.9, 75.2, 74.0, 73.6, 73.2, 72.8,
71.4, 71.3, 69.3, 68.5, 64.5, 60.7, 60.1, 54.7 (2C), 38.5, 27.9,
26.7, 26.2, 25.9; minor stereoisomer, same as major except for
δ 155.8, 109.2, 26.8. MS (PDMS) m/z 1032 (M + Na).
Ack n ow led gm en t. We thank B. O. Pedersen and
K. B. Simonsen for mass spectra.
Su p p or tin g In for m a tion Ava ila ble: NMR peak assign-
ments and 13C NMR spectra of compounds 3-4, 6-7, 9-11,
and 13-23 (25 pages). This material is contained in libraries
on microfiche, immidiatly follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
((1,2:4,5-Diisop r op ylid en efr u ctop yr a n os-3-yl)oxy)((2-
m eth oxyben zyl)oxy)((1-O-m eth yl-2,4,6-tr i-O-ben zyl-R-D-
J O970317Q