Efficient asymmetric synthesis of silanediol precursors from 1,5-dihydrosiloles

Article abstract of DOI:10.1021/ol7021559

Dihydrosiloles are easily prepared from 1,3-dienes and dichlorosilanes, even on kilogram scale. Asymmetric hydroboration of a 3-alkyl-1,5-dihydrosiloie, prepared from a 2-alkyl-1,3-diene, followed by treatment with aqueous HF results in Peterson fragmentation, forming optically active 3-alkyl-4- fluorosilyl-1-butenes. The fluorosilanes are stable to moisture but very reactive toward nucleophiles. In addition, they can be converted to nucleophilic silyllithium reagents.

Full text of DOI:10.1021/ol7021559

ORGANIC
LETTERS
2007
Vol. 9, No. 24
4963-4965
Efficient Asymmetric Synthesis of
Silanediol Precursors from
1,5-Dihydrosiloles
Sushmita Sen, Madhusudhan Purushotham, Yingmei Qi, and
Scott McN. Sieburth*
Department of Chemistry, Temple UniVersity, 1901 North 13th Street, Philadelphia,
PennsylVania 19122
scott.sieburth@temple.edu
Received September 2, 2007
ABSTRACT
Dihydrosiloles are easily prepared from 1,3-dienes and dichlorosilanes, even on kilogram scale. Asymmetric hydroboration of a 3-alkyl-1,5-
dihydrosilole, prepared from a 2-alkyl-1,3-diene, followed by treatment with aqueous HF results in Peterson fragmentation, forming optically
active 3-alkyl-4-fluorosilyl-1-butenes. The fluorosilanes are stable to moisture but very reactive toward nucleophiles. In addition, they can be
converted to nucleophilic silyllithium reagents.
Preparative methods for the synthesis of enantiomerically
enriched organosilane intermediates have provided useful
platforms from which to reach challenging targets and
synthetic goals.1 As part of our continuing exploration of
silanediol-based protease inhibitors 1a, Scheme 1, we have
sought simple and economical approaches to the two
asymmetric silicon substituents inherent in the structure,
R-amino silanes and â-silyl acids.2 We describe here a
straightforward and scalable assembly of optically active
R-alkyl-â-silyl acids.
Scheme 1. Methods for Silanediol Protease Inhibitor
Synthesis
Our first method for â-silyl acid synthesis used an
organolithium reagent carrying the chiral center, 2. While
effective, this approach suffered from length, cost, and
scalability.3 We describe here a general procedure that can
be performed on large scale and that produces the moisture-
insensitiVe fluorosilane intermediate 3. Fluorosilane 3, while
much more convenient to handle than a chlorosilane, remains
very reactive toward nucleophiles and can also be converted
to a nucleophilic silyllithium reagent. The starting materials
for preparation of 3 are the commodity reagent dichlo-
rodiphenylsilane, 4, and 2-alkyl-1,3-diene, 5.
set of dialkyl and diaryl dichlorosilanes and can proceed by
two different mechanisms. An initial complex between
magnesium and the 1,3-diene will react with dichlorodiphe-
nylsilane to give 1,5-dihydrosilole, 6.4 Alternatively, an
intermediate silylene can add to the 1,3-diene.5 We have
limited our investigation to diphenylsilane, because our
synthesis of functionalized silanediols is predicated on the
acidic hydrolysis of diphenylsilanes (Scheme 1).2,3b The
The cornerstone of this chemistry is the magnesium-
mediated one-step “cycloaddition” of 1,3-dienes with dichlo-
rosilanes, Scheme 2. This reaction is effective with a broad
10.1021/ol7021559 CCC: $37.00
© 2007 American Chemical Society
Published on Web 10/30/2007

philic attack of fluoride on the silicon.9 In principle, acidic
cleavage of the phenyl-silicon bonds could compete with
ring cleavage, and could also complicate the product isolation
by replacement of phenyl by additional fluoride. Nonetheless,
the stability of the phenyl silicon bond in 7 is sufficient to
be noncompetetive with the fragmentation. It is likely that
that electronegative fluorine in 3 attenuates the nucleophi-
licity of the phenyl groups and thereby limits further
reactivity under the acidic reaction conditions. The clean and
reproducible HF reaction of 7 contrasts with extension of
this reaction to the corresponding chlorosilane, a reaction
that is highly solvent dependent as well as plagued by
elimination and substitution reactions, and will be described
elsewhere.
Scheme 2. Formation and Fragmentation of Dihydrosilole 6b
The fluorine-silicon bond, one of the strongest covalent
bonds, endows fluorotrialkyl(aryl)silanes with stability to
water under acidic and neutral conditions. This stability,
however, does not preclude their reactivity with nucleophiles.
Indeed, more than 50 years ago Eaborn reported that, for
the synthesis of sterically congested silanes, fluorosilanes
are more reactive electrophiles than the corresponding
chlorosilanes.10
A variety of nucleophiles react with fluorosilane 3, such
as lithiated Boc-protected pyrrolidine that yields silane 9,
Scheme 3. Similarly, metalated thioanisole and 2-lithio-1,3-
dithiane give adducts 10 and 11 in good yield. Even the rather
sensitive chloromethyllithium gave the chloromethyl product
12 in high yield.
magnesium-diene complex can be prepared using magne-
sium turnings, powder, or Reike’s magnesium and can be
formed either prior to addition of the dichlorosilane or in
the presence of the dichlorosilane. A detailed literature
procedure for the preparation of 0.33 kg of distilled 6a using
a single 6-L flask is indicative of the utility of this chemistry.6
We have routinely prepared 6b from isoprene 5b on 20-g
scale. Hydroboration of this alkene yields the trans-alcohol
7 in good yield.
Alcohol 7, when dissolved in a mixture of ethanol and
commercial 48% HF and heated to reflux, forms fluorosilane
3 as the only observed product, which can be isolated by
extraction and purified by distillation or by standard flash
column chromatography in 90% yield.7 This reaction is
essentially a Peterson olefination8 that we envision proceed-
ing through protonated alcohol 8, fragmenting by nucleo-
Scheme 3. Fluorosilane as an Electrophile
(1) (a) Carter, M. J.; Fleming, I.; Percival, A. J. Chem. Soc., Perkin Trans.
1 1981, 2415-2434. (b) Ahmar, M.; Duyck, C.; Fleming, I. Pure Appl.
Chem. 1994, 66, 2049-2052. (c) Beresis, R. T.; Solomon, J. S.; Yang, M.
G.; Jain, N. F.; Panek, J. S. Org. Synth. 1998, 75, 78-88. (d) Lowe, J. T.;
Panek, J. S. Org. Lett. 2005, 7, 3231-3234. (e) Franz, A. K.; Woerpel, K.
A. Acc. Chem. Res. 2000, 33, 813-820. (f) Peng, Z.; Woerpel, K. A. J.
Am. Chem. Soc. 2003, 125, 6018-6019. (g) Hackman, B. M.; Lombardi,
P. J.; Leighton, J. L. Org. Lett. 2004, 6, 4375-4377. (h) Kubota, K.;
Leighton, J. L. Angew. Chem., Int. Ed. 2003, 42, 946-948. (i) Liu, D.;
Kozmin, S. A. Angew. Chem., Int. Ed. 2001, 40, 4757-4759. (j) Liu, D.;
Kozmin, S. A. Org. Lett. 2002, 4, 3005-3007. (k) Ottosson, H.; Steel, P.
G. Chem. Eur. J. 2006, 12, 1576-1585. (l) Sellars, J. D.; Steel, P. G.;
Turner, M. J. Chem. Commun. 2006, 2383-2387. (m) Marion, F.; Calvet,
S.; Marie, J. C.; Courillon, C.; Malacria, M. Eur. J. Org. Chem. 2006, 453-
562. (n) Huckins, J. R.; Rychnovsky, S. D. J. Org. Chem. 2003, 68, 10135-
10145. (o) Denmark, S. E.; Liu, J. H. C. J. Am. Chem. Soc. 2007, 129,
3737-3744.
(2) Sieburth, S. McN.; Chen, C.-A. Eur. J. Org. Chem. 2006, 311-322.
(3) For other attempts to devise improved procedures for silanediol
synthesis see: (a) Organ, M. G.; Buon, C.; Decicco, C. P.; Combs, A. P.
Org. Lett. 2002, 4, 2683-2685. (b) Anderson, T. F.; Statham, M. A. J.;
Carroll, M. A. Tetrahedron Lett. 2006, 47, 3353-3355.
In addition to acting as a reactive electrophile, fluorosilane
3 can be converted to a nucleophilic silyllithium reagent 13,
Scheme 4. While this is not surprising in view of the
comparable use of chlorosilanes in silyllithium synthesis, we
(4) Richter, W. J. Synthesis 1982, 1102. Xiong, H.; Rieke, R. D. J. Org.
Chem. 1989, 54, 3247-3249.
(5) Tortorelli, V. J.; Jones, M., Jr.; Wu, S.; Li, Z. Organometallics 1983,
2, 759-764. Lei, D.; Gaspar, P. P. Organometallics 1985, 4, 1471-1473.
(6) Mignani, S.; Damour, D.; Bastart, J.; Manuel, G. Synth. Commun.
1995, 25, 3855-3861.
(7) Little if any decomposition of alkyl(fluoro)diphenylsilanes are seen
during flash silica gel chromatography; Still, W. C.; Kahn, M.; Mitra, A. J.
Org. Chem. 1978, 43, 2923-2925.
(8) Ager, D. J. Org. React. 1990, 38, 1-223. van Staden, L. F.;
Gravestock, D.; Ager, D. J. Chem. Soc. ReV. 2002, 31, 195-200.
(9) For a closely related reaction involving a six-membered ring, see:
Soderquist, J. A.; Negron, A. Tetrahedron Lett. 1998, 39, 9397-9400. 3,4-
Epoxy tetrahydrosiloles undergo fragmentation when treated with tetrabu-
tylammonium bromide: Park, Y. T.; Manuel, G.; Bau, R.; Zhao, D.; Weber,
W. P. Organometallics 1991, 10, 1586-1591.
(10) Eaborn, C. J. Chem. Soc. 1952, 2840-2846.
4964
Org. Lett., Vol. 9, No. 24, 2007

are not aware of other examples of fluorosilanes being used
for this purpose. Following standard methods (lithium metal,
THF), the conversion of the fluorosilane to a lithium reagent
appears to be more sluggish than a comparable chlorosilane;
the reaction proceeds normally, with the exception of a
precipitation of lithium fluoride.
rane followed by oxidative workup gave alcohol 7. Mosher
ester analysis found the alcohol to have an optical purity of
70% ee, similar to that reported by Brown for the comparable
hydroboration of 1-methylcyclopentene.12 A single recrys-
tallization of 7, however, brought the enantiomeric excess
to at least 95%. Fragmentation of 7 under standard conditions
gave the optically active fluorosilane 3 (Scheme 5).
Scheme 4. Fluorosilane Converted to a Nucleophile
Scheme 5. Asymmetric Preparation of 3
The use of isoprene in the three-step sequence described
here, leads to the chiral 3-methyl-4-(fluorodiphenylsilyl)-1-
butene reagent 3. A broad set of other 1,3-dienes have been
shown to participate in the 1,5-dihydrosilole synthesis that
constitutes the first step of this sequence.13 Each of the three
steps can be easily performed in multigram scale, and no
chromatography is required, although all intermediates are
stable to chromatography. The chiral product 3 is a moisture-
insensitive reagent yet is reactive with a variety of carbon
nucleophiles or can be converted to a moderately nucleophilic
silyl anion. Oxidative cleavage of the alkene provides ready
access to the â-silyl acids that form one-half of the silicon-
based protease inhibitors that provided the incentive for
development of this chemistry.2
Diphenyl substitution of the silicon lowers the nucleophi-
licity of the silyl anion (e.g., dimethylphenylsilyllithium is
more nucleophilic than methyldiphenylsilyllithum). Never-
theless, lithum reagent 13 undergoes addition to cyclohex-
anone to give the R-hydroxysilane 14 in moderate yield. As
originally reported by Scheidt,11 diphenylsilyllithium reagents
are only marginally reactive enough to form adducts with
the Davis sulfoximine 15. While the yield for adduct 16 is
low, only two diastereomers were detectable, suggesting that
diastereocontrol by the sulfoximine group was high. A higher
yield for the silyllithium addition was found with the more
reactive, albeit achiral, phosphonimine 17.
Acknowledgment. Support by the NIH for this program
is gratefully acknowledged. Wondwossen Arasho, Swapnil
Singh, and Nishith Sanghvi contributed to the chemistries
supporting the work described here.
Supporting Information Available: Experimental pro-
cedures, characterization data, and 1H NMR spectra of new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL7021559
Fluorosilane 3 can also be prepared in optically active
form. Treatment of alkene 6b with mono-isopinocamphylbo-
(12) Brown, H. C.; Schwier, J. R.; Singaram, B. J. Org. Chem. 1978,
43, 4395-4397. Brown, H. C.; Jadhav, P. K.; Mandal, A. K. J. Org. Chem.
1982, 47, 5074-5083.
(11) Ballweg, D. M.; Miller, R. C.; Gray, D. L.; Scheidt, K. A. Org.
Lett. 2005, 7, 1403-1406.
(13) Rieke, R. D.; Xiong, H. J. Org. Chem. 1991, 56, 3109-3118.
Org. Lett., Vol. 9, No. 24, 2007
4965