Stereoselective oxy-functionalization of γ-silyl allylic alcohols with ozone: A facile synthesis of silyl peroxide and its reactions

Article abstract of DOI:10.1021/ol061635r

A reaction of γ-silyl allylic alcohol and its ether with ozone provides synthetically versatile α-formyl silyl peroxide in good yield without normal fission of carbon-carbon double bond. Thus, the provided silyl peroxide serves as a good precursor for the

Full text of DOI:10.1021/ol061635r

ORGANIC
LETTERS
2006
Vol. 8, No. 18
4023-4026
Stereoselective Oxy-Functionalization of
-Silyl Allylic Alcohols with Ozone: A
Facile Synthesis of Silyl Peroxide and
Its Reactions
γ
Masanori Murakami, Kyohei Sakita, Kazunobu Igawa, and Katsuhiko Tomooka*
Department of Applied Chemistry, Graduate School of Science and Engineering, Tokyo
Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan
ktomooka@apc.titech.ac.jp
Received July 3, 2006
ABSTRACT
A reaction of
γ
-silyl allylic alcohol and its ether with ozone provides synthetically versatile
r-formyl silyl peroxide in good yield without
normal fission of carbon
−carbon double bond. Thus, the provided silyl peroxide serves as a good precursor for the stereochemically defined
triol derivative via alkylation and reduction of peroxide moiety.
The addition of hydroxy groups or their equivalents to the
carbon-carbon double bond is a fundamental and valuable
transformation in organic synthesis, and numerous methods
to effect this transformation have been developed.1 However,
further development of this area is necessary to respond to
the demand of modern fine organic synthesis. Herein, we
report an efficient and unique oxyfunctionalization of the
alkene moiety of γ-silyl allylic alcohol i using ozone, which
provides synthetically versatile R-formyl silyl peroxyalde-
hyde ii without the normal fission of the carbon-carbon
double bond. Thus, the provided ii serves as a good precursor
for the stereochemically defined triol derivative iii via
alkylation and reduction of peroxide moiety.
OsO4, m-CPBA, etc. Although most methods did not give
any significant results, the reaction of 1a (i: R1, R2 ) PhC2H4,
R3Si ) t-BuMe2Si) with ozone in CH2Cl2 proceeded
smoothly to give unexpected dihydroxy silyl peroxide 2a
(30%) in addition to the normal ozonolysis product 3a (26%)
after a reductive workup using NaBH4.3-8 In addition, the
silyl peroxide 2a was obtained exclusively in 75% yield by
changing the solvent from CH2Cl2 to AcOEt.9,10
To explore the scope and limitation of this unique
ozonation, we examined a similar reaction using a variety
of substrates as shown in Table 1.11 Again, in a reaction of
(1) For leading reviews on dihydroxylation of carbon-carbon double
bonds, see: (a) Schro¨eder, M. Chem. ReV. 1980, 80, 187-213. (b) Haines,
A. H. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon: Oxford, 1991; Vol. 7, pp 437-448. For leading reviews on
epoxidation of carbon-carbon double bond, see: (c) Jo¨rgensen, K. A. Chem.
ReV. 1989, 89, 431-458. (d) Rao, A. S. In ComprehensiVe Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol.
7, pp 357-387.
We found this oxidation during the course of our study
on hydroalumination of γ-silyl propargylic alcohol providing
γ-silyl allylic alcohol i.2 To explore the synthetic utility of
i, we planned to carry out the oxyfunctionalization of the
vinyl silane moiety using other common oxidants such as
10.1021/ol061635r CCC: $33.50
© 2006 American Chemical Society
Published on Web 08/10/2006

allylic alcohols 1c and d (entries 3 and 4). Moreover, silyl
peroxides were also obtained upon ozonation of the corre-
sponding trimethylsilyl ethers 1e and f, indicating that a free
hydroxyl group is not essential for the reaction (entries 5
and 6). Primary ether 1g also gave silyl peroxide 2g in
excellent yield (entry 7). We have observed diastereoselective
addition of ozone onto substrates 1h-k which bear an
R-stereogenic center. The anti-selectivity is more pronounced
as the R1 group becomes more bulky. In particular, a reaction
of 1k (R1 ) t-Bu) gave anti-2k exclusively (>99% dr) (entry
11).13,14 It is worth noting that this oxidation was applicable
to the γ-silyl allylic alcohol 1l having an alkynyl substituent
(entry 12). These results clearly show that the present
oxidation provides an efficient approach to the well-
functionalized silyl peroxides, in which otherwise difficult
triorganosilicon-sp2 carbon bond cleavage may be involved.15
Since organic peroxides are potentially hazardous com-
pounds, they must be handled with due care.16 However, no
particular difficulties were experienced in handling any of
the new silyl peroxides synthesized in this work. Also, we
have observed thermal stability of 2b and 4e by thermo-
gravimetry analysis (TGA) under nitrogen. It shows that the
slow thermal degradation occurred at >80 °C (Figure 1).
Table 1. Oxidation of γ-Silyl Allylic Alcohol and Its Ether 1
yield
workup
proced-
urea
(%)b
dr (anti/
syn)c
entry
1
R1
R2
X
R3Si
2
4
1
2
3
4
5
6
7
8
9
1b Me
1b Me
1c Me
1d Me
1e Me
1f Me
Me H
Me H
Me H
Me H
t-BuMe2Si
t-BuMe2Si
i-Pr3Si
A
B
B
A
B
B
A
A
A
A
A
A
73
nd
nd
70
nd
nd
90
nd
73
80
nd
75
67
nd
t-BuPh2Si
Me TMS i-Pr3Si
Me TMS t-BuMe2Si
1g
H
H
H
H
H
H
TBS t-BuPh2Si
TMS i-Pr3Si
TMS i-Pr3Si
TMS i-Pr3Si
TMS i-Pr3Si
1h Me
1i Et
78 58:42 nd
84 67:33 nd
89 72:28 nd
71 >99:<1 nd
90 74:26d nd
10 1j c-Hex
11 1k t-Bu
12 1l i-Pr3 Si9t Me H
i-Pr3Si
a A: Reductive workup using NaBH4. B: Reaction was worked up
without any reducing agent. b Isolated yields. nd ) not detected by 1H NMR
and TLC analyses. c Determined by 1H NMR analysis. d Stereochemistry
of the major product was not determined yet.
1b, AcOEt was the solvent of choice to obtain 2b in good
yield (entry 1). Furthermore, R-formyl silyl peroxide 4b was
isolated in good yield by concentration of the crude mixture
under reduced pressure (entry 2).12 Equally good yields of
silyl peroxides were obtained from γ-i-Pr3Si or γ-t-BuPh2Si
(2) Igawa, K.; Tomooka, K. Angew. Chem., Int. Ed. 2006, 45, 232-
234.
(3) For example, both reactions of 1a with OsO4 or m-CPBA at rt did
not proceed, only the starting material being recovered.
(4) To confirm the structure of silyl peroxide 2a, we compared it with
an authentic sample of TBS ether which had been prepared from diben-
zylideneacetone in five steps ((i) H2, Pd/C, (ii) CH2dCHMgBr, (iii) OsO4,
N-methylmorphorin N-oxide, (iv) t-BuMe2SiCl, imidazole, (v) H3O+). On
13C NMR analyses, the â-carbon of 2a appears in much lower field (88.9
ppm) than that of TBS ether (75.7 ppm); see the Supporting Information.
(5) For leading reviews on ozonolysis, see: (a) Bailey, P. S. Ozonation
in Organic Chemistry, Vol. 1: Olefinic Compounds; Academic Press:
London, 1978. (b) Bailey, P. S. Ozonation in Organic Chemistry, Vol. 2:
Nonolefinic Compounds; Academic Press: London, 1982.
Figure 1. TGA of 2b and 4e.
(6) Bu¨chi and Wu¨est reported the ozonation of trimethylsilyl-substituted
alkenes in the 1970s, in which they proposed a similar silyl peroxide as an
intermediate; see: Bu¨chi, G.; Wu¨est, H. J. Am. Chem. Soc. 1978, 100, 294-
295. Although their trimethylsilyl peroxide is too reactive to be handled
with ease, our t-BuMe2Si peroxide 2a is tolerant not only to the reductive
workup process using NaBH4 but also to purification on silica gel, most
probably due to the bulky silyl group on the peroxide moiety.
(7) Synthetic application of ozonation of vinylsilane is rather limited,
so far. For references on synthesis of artemisinin, see: (a) Avery, M. A.;
Chong, W. K. M.; Jennings-White, C. J. Am. Chem. Soc. 1992, 114, 974-
979. For a reference on the synthesis of R-hydroxyketone, see: (b) Renaud,
P.; Gerster, M.; Ribezzo, M. Chimia 1994, 48, 366-369.
To gain information on the mechanism of this oxidation,
the reaction of 1b with ozone at -78 °C in hexane was
monitored by IR spectroscopy (Figure 2).17,18 When ozone
was passed through a substrate solution with a constant flow,
(12) R-Formyl silyl peroxide was utilized as a precursor of peroxide
hemiketal; see: Clark, G. R.; Nikaido, M. M.; Fair, C. K.; Lin, J. J. Org.
Chem. 1985, 50, 1994-1996.
(8) 1,2-Diol synthesis via the ozonation of alkenylstannane has been
reported; see: Go´mez, A. M.; Company, M. D.; Valverde, S.; Lo´pez, J. C.
Org. Lett. 2002, 4, 383-386.
(13) The stereochemistry of 2h-k was determined by the 1H NMR
analysis of their cyclic acetal derivatives; see the Supporting Information.
(14) Typical Procedures of Ozonation (Procedure B). A stream of
ozone (1.2 v/v % in oxygen, 150 mL/min) was bubbled through a solution
of allylic alcohol 1e (747 mg, 2.38 mmol) in AcOEt (30 mL) at -78 °C.
After 1 h, the solution turned pale blue, indicating the complete oxidation.
Dissolved ozone was removed by bubbling the solution with argon for 15
min followed by allowing the temperature to rise to rt. After the solvent
was removed by evaporation, the residue was purified by silica gel
chromatography (hexane: Et2O ) 100:1) to afford silyl peroxide 4e (644
mg, 1.78 mmol, 75%) as a colorless oil.
(9) The reaction in MeOH instead of AcOEt only afforded decomposed
materials resulting from a concomitant retro-aldol reaction.
(10) For representative references on silyl peroxide, see: (a) Isayama,
S.; Mukaiyama, T. Chem. Lett. 1989, 573-576. (b) Dussault, P. H.; Lee,
I. Q.; Lee, H.-J.; Lee, R. J.; Niu, Q. J.; Schultz, J. A.; Zope, U. R. J. Org.
Chem. 2000, 65, 8407-8414. (c) O’Neill, P. M.; Pugh, M.; Davies, J.; Ward,
S. A.; Park, B. K. Tetrahedron Lett. 2001, 42, 4569-4571. (d) Tokuyasu,
T.; Kunikawa, S.; Masuyama, A.; Nojima, M. Org. Lett. 2002, 4, 3595-
3598 and references therein.
(11) All of the substrates 1 were readily available by the hydroalumination
of the corresponding propargylic alcohol using Red-Al in toluene.
(15) For review on the oxidation of the silicon-carbon bond, see: Jones,
G. R.; Landais, Y. Tetrahedron 1996, 52, 7599-7662.
4024
Org. Lett., Vol. 8, No. 18, 2006

suitable, in which the R3Si group migrates from carbon to
oxygen more rapidly than a normal ozonide rearrangement
(7 f 8).23,24
Our attention was next directed toward further transforma-
tion of silyl peroxides. The reaction of 4e with alkyllithiums
in toluene gave syn-9 (racemic) with the preservation of
peroxide moiety with good to excellent diastereoselectivity
(Scheme 1).25,26 We also found that the chiral silyl peroxides
Scheme 1. Transformation of 4e
Figure 2. In situ IR of the reaction of 1b with ozone.
we observed a time-dependent diminution in the intensity
of the CdC absorption of 1b (1616 cm-1) with the
emergence of CdO absorption of silyl peroxide 4b (1738
cm-1). This clear correlation between decreasing of 1b and
the increasing of 4b and the absence of significant absorption
for any species other than 1b or 4b in this functional group
region suggests that the reaction proceeds rapidly via a short-
lived intermediate.19
Although the present reaction could proceed via interme-
diacy of epoxide 520 or silylenol ether 6,21 a control
experiment showed that separately prepared epoxide 5a (R3-
Si ) i-Pr3Si, R′ ) C(OH)MeCtCH) or enol ether 6a (R3Si
) t-BuMe2Si, R′ ) C(OTBS)Me2) did not give the corre-
sponding silyl peroxides.22
are convertible to the corresponding mono-deoxygenated
products in a stereospecific manner. For example, a reduction
of syn-10 using Me3P at rt yielded syn-11 without loss of
diastereopurity.27
Moreover, these stereoselective transformations along with
a diastereoselective formation of silyl peroxide (vida supra)
enable the stereoselective construction of three continuous
chiral centers in (1,2-anti, 2,3-syn)-13 as shown in Scheme
Scheme 2. Constraction of Three Continuous Chiral Centers
On the basis of the above-mentioned results, we believe
that the primary ozonide pathway (7 f 4) is the most
(16) (a) Castrantas, H. M.; Banerjee, D. K.; Noller, D. C. Fire and
Explosion Hazards of Peroxy Compounds, ASTM STP 394; American
Society for Testing and Materials: Philadelphia, PA, 1965. (b) Castrantas,
H. M.; Banerjee, D. K. Laboratory Handling and Storage of Peroxy
Compounds, ASTM STP 471; American Society for Testing and Materials:
Philadelphia, PA, 1970.
(17) In situ IR spectra were recorded on MCT detector fitted with the
mid-IR fiber-optic probe (Remspec). The spectra were acquired using 10
scans per spectrum at a resolution of 4 cm-1 using Time Base software
(Perkin Elmer).
(18) AcOEt was not employed as a solvent in this experiment due to its
large background absorption in IR spectroscopy.
(19) Oxidation in hexane provides not only a silyl peroxide, but also a
small amount of normal ozonolysis product.
(20) It has been reported that an epoxide can be formed in ozonation of
alkene; see: (a) Gillies, C. W. J. Am. Chem. Soc. 1975, 97, 1276-1278.
(b) Bailey, P. S.; Hwang, H. H.; Chiang, C. Y. J. Org. Chem. 1985, 50,
231-234 and references therein.
(21) It has been reported that an R-silyloxy ketone was provided in
ozonation of silyl enol ether; see: Clark, R. D.; Heathcock, C. H. J. Org.
Chem. 1976, 41, 1396-1403.
(22) The reaction of epoxide 5a and O3/O2 gas did not give any oxidized
products. Under the same conditions, silyl enol ether 6a gave a complex
mixture including no silyl peroxide.
2. The intermediary silyl peroxide 12 was also convertible
to the R,R′-dihydroxyketone derivative anti-14 upon base
(23) A similar silyl migration mechanism was speculated by Bailey to
explain Bu¨chi’s result; see ref 5b. As a part of a mechanistic study for
ozonolysis of alkene, Pola and colleagues investigated an oxidation of
trimethylvinylsilane in a cryogenic system. They proposed that the silyl
peroxide was formed from primary ozonide via homolytic cleavage of the
oxygen-oxygen bond; see: Fajgar, R.; Roithova´, J.; Pola, J. J. Org. Chem.
2001, 66, 6977-6981.
Org. Lett., Vol. 8, No. 18, 2006
4025

treatment. These results clearly show that the silyl peroxide
functionality acts as a synthetic equivalent of silyl ether or
masked carbonyl.28
In summary, we have described an oxidation of γ-silyl
allylic alcohol using ozone, which provides an access to
synthetically versatile silyl peroxides. Further studies using
chiral-nonracemic allylic alcohols prepared by recently
developed enantioselective hydroalumination protocol in our
laboratory are now underway.29
Acknowledgment. This research was supported in part
by a Grant-in-Aid for Scientific Research on Priority Areas
(A) “Creation of Biologically Functional Molecules”, Ex-
ploratory Research No. 17655038, Young Scientists (B) No.
18750081, and The 21st Century COE Program “Creation
of Molecular Diversity and Development of Functionalities”
from the Ministry of Education, Culture, Sports, Science and
Technology, Japan. We thank Professor Shinji Ando for
technical assistance in TGA measurement.
(24) A related silyl migration in furan-derived peroxide has been reported;
see: (a) Adam, W.; Rodriguez, A. Tetrahedron Lett. 1981, 22, 3505-3508.
(b) Katsumura, S.; Hori, K.; Fujiwara, S.; Isoe, S. Tetrahedron Lett. 1985,
26, 4625-4628.
(25) The stereochemistry of 9-14 was determined through the 1H NMR
analysis of their cyclic acetal derivatives; see the Supporting Information.
(26) This stereochemical outcome is explicable on the basis of chelation-
controlled mechanism, in which the lithium cation coordinates to carbonyl
oxygen and R-silyl oxygen.
Supporting Information Available: Experimental pro-
cedures and spectral data. This material is available free of
charge via the Internet at http://pubs.acs.org.
(27) This result means that the R-silyl oxygen of peroxide was selectively
removed in the reduction. A related reduction of silyl peroxide in a
norcamphor derivative has been reported; see: Jefford, C. W.; Rimbault,
C. G. J. Am. Chem. Soc. 1978, 100, 6437-6445.
(28) Cyclic peroxide has been utilized as a synthetic equivalent of ketone;
see: Singh, C.; Malik, H. Org. Lett. 2005, 7, 5673-5676.
OL061635R
(29) These results will be reported in due course.
4026
Org. Lett., Vol. 8, No. 18, 2006