Tris(pentafluorophenyl)silyl triflate: Synthesis and silylation of carbonyl compounds

Article abstract of DOI:10.1002/ejoc.200400552

Tris(pentafluorophenyl)silyl triflate (1) was prepared by protodesilylation of phenyl-, allyl-, and isopropenyloxytris(pentafluorophenyl)silyl derivatives and characterized by X-ray crystallography. This reagent was employed for the silylation of carbonyl compounds. Aldehydes and ketones afforded the corresponding silyl enol ethers in good yields, while silylation of esters and lactones was dependent on the reaction conditions and on the substrate. A mechanism accounting for the observed phenomena is proposed. Studies of the relative reactivities of 1 and trimethylsilyl triflate are also presented. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejoc.200400552

FULL PAPER
Tris(pentafluorophenyl)silyl Triflate: Synthesis and Silylation of Carbonyl
Compounds
Vitalij V. Levin,[a] Alexander D. Dilman,*[a] Pavel A. Belyakov,[a] Alexander A. Korlyukov,[b]
Marina I. Struchkova,[a] and Vladimir A. Tartakovsky[a]
Keywords: Carbonyl compounds / Perfluorinated ligands / Silicon / Silylation
Tris(pentafluorophenyl)silyl triflate (1) was prepared by pro-
ation of esters and lactones was dependent on the reaction
todesilylation of phenyl-, allyl-, and isopropenyloxytris(pen- conditions and on the substrate. A mechanism accounting for
tafluorophenyl)silyl derivatives and characterized by X-ray
crystallography. This reagent was employed for the silylation
of carbonyl compounds. Aldehydes and ketones afforded the
corresponding silyl enol ethers in good yields, while silyl-
the observed phenomena is proposed. Studies of the relative
reactivities of 1 and trimethylsilyl triflate are also presented.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
facile extrusion of sulfur trioxide driven by the
siliconϪfluorine bond formation. Nevertheless, one can ex-
pect that the analogous TPFS triflate, (C6F5)3SiOSO2CF3
(1), will be more stable since it has no obvious degra-
dation pathway.
Introduction
Silylating agents of general formula R3SiX have found
widespread use in synthetic chemistry for the modification
of functional groups[1] and as Lewis-acidic mediators in
carbonϪcarbon bond-forming reactions.[2] The primary
factor determining the strength of a silylating species is the
nature of the leaving group X as this affects the electrophil-
icity of the silicon atom. Although many compounds with
different leaving groups have been described,[3] derivatives
of triflic acid, such as silyl triflates, have turned out to be
the most popular reagents for a wide variety of appli-
cations.[2,4] The reactivity of silyl triflates can be further
controlled by varying the steric volume of the alkyl sub-
stituents. At the same time it would be interesting to con-
sider carbon-centered groups which, besides being simple
spectators, could also exert an electronic effect. Governed
by this idea we decided to examine a silylating reagent bear-
ing three pentafluorophenyl fragments at the silicon atom.[5]
Recently, Frohn has attempted to prepare tris(pentafluor-
ophenyl)silyl fluorosulfonate (TPFS-OSO2F), (C6F5)3-
SiOSO2F; he found this compound to be unstable as it de-
composes into the corresponding fluorosilane (C6F5)3SiF.[6]
This instability is probably associated with the
The use of triflate 1 as a silylating reagent will allow the
introduction of the TPFS fragment into a substrate. We
have previously used TPFS chloride for the silylation of ke-
tones.[7] This reaction was performed in refluxing dichloroe-
thane and afforded TPFS enol ethers in moderate yields.
The possibility of using TPFS enol ethers in aldol reac-
tions[7] further justifies work to improve the procedure for
this synthesis. The application of triflate 1 for the silylation
of carbonyl compounds is expected to solve the latter prob-
lem, since in silylation reactions silyl triflates usually give
better results than the corresponding chlorosilanes.[4a,4b]
In this paper we report the synthesis and characterization
of silyl triflate 1 and demonstrate its efficiency in the silyl-
ation of aldehydes, ketones, and esters.
Results and Discussion
Synthesis of TPFS Triflate
The interaction of organosilanes with triflic acid is the
most convenient way to form silyl triflates.[4a,4b] Based on
extensive literature data, the propensity of a substituent to
undergo protodesilylation should increase in the following
order:[4a,4b,8] vinyl Ͻ phenyl Ͻ allyl Ͻ isopropenyloxy. We
investigated the protodesilylation of several TPFS deriva-
tives (Scheme 1).
[a]
N. D. Zelinsky Institute of Organic Chemistry
119991 Moscow, Leninsky prosp. 47, Russian Federation
Fax: ϩ7-095-135-5328
E-mail: adil25@mail.ru
A. N. Nesmeyanov Institute of Organoelement Compounds,
119991 Moscow, Vavilov str. 28, Russian Federation
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
[b]
Eur. J. Org. Chem. 2004, 5141Ϫ5148
DOI: 10.1002/ejoc.200400552
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5141

A. D. Dilman et al.
FULL PAPER
inexpensive bromopentafluorobenzene and phenyltrichloro-
silane,[6] triflate 1 may be considered as readily available.
X-ray Structure of 1
The molecular structure of TPFS triflate was investigated
by X-ray diffraction analysis (Figure 1). Analysis of the ge-
ometry of 1 showed that the silicon atom has an almost
undistorted tetrahedral configuration. Of particular note is
˚
the length of the SiϪO bond (1.663 A), which is consider-
Scheme 1
ably shorter than that in analogous triphenylsilyl triflate
(1.742 A), and is the shortest SiϪO distance known
˚
amongst other published silyl triflates.[9] A decrease of the
SiϪO bond length was also observed for the structure of
the TPFS enol ether 3c,[7] and seems to be a general
phenomenon for TPFS derivatives. It may be due to the
influence of the electron-withdrawing pentafluorophenyl
groups.
A very slow reaction was observed between vinylsilane
and TfOH, leading to less than 10% conversion in 24 hours.
It is worthy to note that protodesilylation of triethylvinylsil-
ane occurs instantaneously under the same conditions. Such
a significant difference between TPFS and trialkylsilyl sys-
tems is apparently associated with the strong electron-with-
drawing effect of the pentafluorophenyl groups.
The reaction of phenyl-TPFS with one equivalent of acid
proceeded at room temperature in chlorinated solvents.
However, as the conversion increased the rate of the process
decreased dramatically, furnishing samples of triflate 1 that
were always contaminated with the starting phenylsilane.
We conjectured that TfOH is deactivated by complexation
with the triflate 1, but we could not observe any complex
formation by 1H and 19F NMR spectroscopy. Complete
conversion was achieved by using either 2.1 equivalents of
TfOH at room temperature or with 1.6 equivalents and
mixing the reagents at 60 °C in dichloroethane. In both
cases the yield of TPFS triflate 1, after purification by
recrystallization from dichloroethane, was between 83 and
92%, although the latter procedure, which requires less acid,
is more practical for larger scale syntheses.
Upon interaction of allyl-TPFS with one equivalent of
TfOH in CDCl3 the starting silane was consumed quite rap-
idly to afford moderate quantities of triflate 1 along with
small amounts of isopropyl triflate and other unidentified
species. The same reaction with 2.1 equivalents of TfOH
gave a stoichiometric mixture of triflate 1 and isopropyl
triflate, as judged by NMR spectroscopy, while a prepara-
tive run performed in dichloromethane provided triflate 1
in 73% yield.
Figure 1. The molecular structure of TPFS triflate 1; atoms are
shown as thermal ellipsoids at 50% probability; selected bond
˚
lengths (A): Si(1)ϪO(1) 1.663(2), SiϪC (average) 1.858(3), SϪO(1)
(average) 1.510(2), SϪO 1.433(3); bond angles (°) SiϪOϪS (aver-
age) 149.5, OϪSiϪC (average) 106.8, CϪSiϪC 112.0
Silyloxypropene (3c), which is believed to be the most
active towards acid, reacts instantaneously with 1.1 equiva-
lents of TfOH to afford only triflate 1 and acetone, accord-
ing to NMR spectroscopy. The preparative experiment gave
a yield of 87% of the desired triflate.
Silylation of Aldehydes and Ketones
Thus, triflate 1 can be accessed from several TPFS pre-
Various carbonyl compounds can be silylated by triflate
cursors. Although the silyl enol ether gives shorter reaction 1 in the presence of triethylamine in dichloromethane
times, the need to prepare it[7] renders this approach some- [Equation (1), Table 1]. The silylation is usually complete
what tedious. From a practical point of view, protodesilyl- within 30 minutes at room temperature, although sterically
ation of the phenylsilane is recommended. Taking into ac- hindered substrates may require slightly longer reaction
count that the phenylsilane itself is easily synthesized from times (entries 5 and 11)
5142
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 5141Ϫ5148

Synthesis and Silylation of Carbonyl Compounds
FULL PAPER
Table 1. Synthesis of TPFS enol ethers 3
Entry
2
R1
R2
Time
(h)
3
Yield of 3 E:Z
(%)[a]
1
2
3
4
5
6
7
8
9
2a
2b
2c
2d
2e
H
H
H
Me
H
H
H
H
Me
Me
0.5
0.5
0.5
0.5
1
0.5
0.5
0.5
0.5
0.5
3
3a
3b
3c
3d
3e
3f
72
82
92
88
93
92
87
89
94
95
1:10[b]
Me
Ph
tBu
2f[c] cC3H5
2g
2h
2i
Et
Ph
Ϫ(CH2)4Ϫ
Ϫ(CH2)5Ϫ
3g
3h
3i
1:3.3[d]
1:8.5[d]
10
11
2j
3j
2k[e] (ϩ)-camphor
3k 86
[a]
[b]
Yield of distilled product.
Configuration was determined by
comparison of H-H coupling constants in the HCϭCH fragment.
[c] Methyl cyclopropyl ketone was used. [d] Configuration was deter-
[e]
mined from the NOE spectrum. (ϩ)-Camphor was used.
(1)
Scheme 2
Since triflate 1 is moderately soluble and products 3 are
usually well soluble in dichloromethane, the progress of the
reaction can be followed by disappearance of the silylating
agent. However, enol ether 3c has limited solubility in di-
chloromethane and may precipitate on the surface of the
triflate 1, thereby preventing further dissolution of the latter
and blocking the reaction. Fortunately, this problem can
easily be overcome by using slightly more solvent.[10]
The silylation of substrates 2b,g,h, all of which possess
an ethyl group at the carbonyl carbon, afforded preferen-
tially Z-isomers. Such a phenomenon has been observed for
silylations with other silylating reagents and may originate
from the greater thermodynamic stability of Z-isomers.[11]
This stabilizing effect presumably comes from the interac-
tion between the σ(CϪH) and σ*(CϪO) orbitals across the
double bond, which are anti-oriented only in Z-configured
compounds.
lyl enol ether 3n was obtained on silylation of 2n using Hu-
nig’s base in refluxing toluene. Obviously, the steric hin-
drance of the base and high temperature favor the dis-
sociation of the ammonium salt back to the ketone and the
triflate 1.
The use of triflate 1 for the silylation of carbonyl com-
pounds offers significant advantages over TPFS chloride in
terms of reaction times, temperature, as well as product
yield and purity. Silylations with 1 usually proceed quite
cleanly to give crude products of good purity that contain
only traces of the starting substrate and triethylammonium
triflate. If desired, TPFS enol ethers 3 can be distilled in
vacuo.[13] In contrast to trialkylsilyl enol ethers, compounds
3 are extremely moisture-sensitive species that revert back
to carbonyls on hydrolysis. While problematic at first
glance, TPFS enol ethers can be conveniently stored and
handled using traditional Schlenk techniques.
The silylation of 3-methylbutan-2-one (2l) under stand-
ard conditions afforded the silyl enol ether 3l-term along
with small amounts of the internal regioisomer (a in
Scheme 2). Lowering the temperature, changing the solvent
Silylation of Esters and Lactones
The silylation of methyl acetate under standard con-
from dichloromethane to toluene, and performing the ad- ditions afforded α-silyl ester 5 as the sole product
dition of a solution of triflate 1 to the mixture of ketone (Scheme 3). NMR monitoring of this reaction revealed that
and triethylamine allowed us to isolate the terminal isomer the silyl ketene acetal 6 is initially formed, followed by re-
3l-term as the only detectable product. Disappointingly, the arrangement into 5.[14] Importantly, the transformation of
silylation of butan-2-one (2m) under different conditions al- 6 to 5 begins when the starting methyl acetate and the silyl-
ways gave approximately a 1:1 mixture of regioisomers (b ating agent have not been completely consumed. This ob-
in Scheme 2).
servation prompted us to propose that the conversion of the
Attempts to silylate methyl vinyl ketone with triflate 1 ketene acetal to the α-silyl ester is promoted by the silylat-
and triethylamine at room temperature failed. According ing agent rather than occurring as a concerted silyl-group
to NMR spectroscopic data, upon mixing the reagents the migration. If this is the case, to generate ketene acetal 6 as a
formation of ammonium salt 4 occurred rapidly instead of major component it is necessary to accelerate its formation,
methyl group silylation[12] (c in Scheme 2). Nevertheless, si- which is most easily achieved by increasing the concen-
Eur. J. Org. Chem. 2004, 5141Ϫ5148
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5143

A. D. Dilman et al.
FULL PAPER
tration of starting substrate. Indeed, performing the silyl-
ation with methyl acetate as solvent at Ϫ78 °C to room
temperature allowed us to isolate ketene acetal 6 as the sole
product. Compound 6 is a somewhat labile species that is
prone to rearrangement into the C-isomer at room tempera-
ture. However, it may be stored at Ϫ25 °C for at least twelve
days with no signs of rearrangement. Thus, depending on
the reaction conditions, it is possible to obtain either the C-
or O-silyl derivatives of methyl acetate.
Figure 2. The molecular structure of 9; non-hydrogen atoms are
shown as thermal ellipsoids at 50% probability; selected bond
˚
lengths (A): Si(1)ϪO(1) 1.632(3), O(1)ϪC(1) 1.347(5), SiϪC (aver-
age) 1.886(5); bond angles (°): Si(1)ϪO(1)ϪC(1), CϪSiϪC 110.14,
OϪSiϪC 108.66
likely that silylation proceeds through the silyloxycarben-
ium cation 10, which is formed upon transfer of the TPFS
group from the silylating agent to the substrate[19]
(Scheme 4). Subsequent deprotonation should take place
from conformation 11, in which breaking of the CϪH bond
occurs parallel to the cationic π-system. Cation 10 may also
exist in four different conformations 10aϪd.
Scheme 3
Silylation of methyl propionate furnished the stable and
distillable silyl ketene acetal 7 as the Z-isomer only.[15] Sur-
prisingly, methyl isobutyrate did not undergo silylation un-
der standard conditions.[16] Silylation of butyrolactone and
α-methylbutyrolactone was quite rapid and clean, leading
to silyl ketene acetals 8 and 9, respectively. The molecular
structure of 9 was studied by X-ray crystallography (Fig-
ure 2). This is the first structural characterization of a
TPFS-ketene acetal. As expected, the SiϪO bond length in
˚
9 (1.632 A) is shorter than that in non-fluorinated deriva-
tives.[17]
Triflate 1 is a more selective reagent than conventional
trialkylsilyl triflates for the silylation of esters and lactones.
Indeed, silylation of esters with trimethylsilyl triflate usually
provides either mixtures of C- and O-silyl derivatives or bis-
silylated species.[4a,4b,18] The latter are also formed from bu-
tyrolactone.[4a,4b,18] In contrast, as shown in the present
study, TPFS triflate always affords single products, and the
problem of bis-silylation was not encountered.
The reactivity difference between methyl isobutyrate and
α-methylbutyrolactone, both of which are α,α-disubstituted
esters, is remarkable. To explain this observation it is neces-
sary to consider the mechanism of the reaction. It is quite Scheme 4
5144  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjoc.org
Eur. J. Org. Chem. 2004, 5141Ϫ5148

Synthesis and Silylation of Carbonyl Compounds
FULL PAPER
To account for the experimental data two hypotheses may very convenient reagent for the introduction of a TPFS
be proposed: group by silylation, as demonstrated by its use for the syn-
Deprotonation occurs from conformation 10a: In the reac- thesis of TPFS enol ethers and ketene acetals. The avail-
tion of methyl isobutyrate this structure is destabilized ow- ability of the latter species provides a further stimulus for
ing to repulsive interaction between the methyl groups (R1, the elaboration of their chemistry. Based on the results pre-
R2 ϭ Me), which becomes even more pronounced in the sented in this paper, two general directions for further inves-
transition state (see projection 11).
tigation may be identified: employment of triflate 1 for the
Deprotonation occurs from conformation 10c: In this case, silylation of functional groups, and utilization of TPFS de-
for methyl isobutyrate the steric repulsion between the α- rivatives in synthetic methodology. Studies along these lines
methyl substituent (R2) and the bulky silyl group is destabil- are in progress in our laboratories.
izing.
Reaction of α-methylbutyrolactone probably proceeds via
the conformation 10a, which is devoid of unfavorable inter-
Experimental Section
actions. By applying this analysis to the silylation of methyl
propionate we may expect that conformer 10a should give
General Remarks: All reactions were performed under an argon
atmosphere. Dichloromethane (DCM) and dichloroethane (DCE)
the Z-isomer, while conformer 10c should give the E-isomer
of the silyl ketene acetal. In practice, (Z)-7 was obtained
exclusively, thereby supporting the first hypothesis implying
10a as the product-determining conformation in the
acyclic series.
were refluxed over anhydrous AlCl3 for two days followed by distil-
lation, then refluxed and distilled from NaOH, and finally distilled
from CaH2 prior to use. Hexane was distilled from LiAlH4 and
˚
stored over 4-A molecular sieves. Toluene was distilled from CaH2
˚
and stored over 4-A molecular sieves. All NMR measurements
were carried out in CDCl3 which was distilled from CaH2 and
˚
stored over 4-A molecular sieves. Melting points were determined
Activity of Triflate 1
in a sealed capillary. The starting compounds TPFS phenyl,[6,22]
TPFS allyl,[22] 3c,[7] and α-methylbutyrolactone[23] were obtained
according to literature procedures.
It is important to evaluate the activity of TPFS triflate 1
in comparison with its trialkylsilyl analogue, since this will
give a measure of the influence of the pentafluorophenyl
group. The propensity of TPFS enol ethers to hydrolyse, as
well as their ability to undergo uncatalyzed coupling with
aldehydes,[7] testifies in favor of the enhanced Lewis acidity
of the TPFS fragment compared to its trialkylsilyl counter-
part. In addition, the acceleration of the nucleophilic attack
at the silicon atom upon the introduction of a pentafluoro-
phenyl group has been documented in earlier investi-
gations.[20] Accordingly, one can expect that TPFS triflate
1 should be more reactive than conventional trimethylsilyl
triflate (TMSOTf, 12).
TPFS Triflate (1). From TPFS Phenyl: Triflic acid (1.12 mL,
12.56 mmol) was added dropwise to a stirred solution of TPFS
phenyl (4.76 g, 7.85 mmol) in DCE (5 mL) at 60 °C. The flask was
wrapped with cotton wool and allowed to stand at room tempera-
ture for 6 hours without stirring, and then was kept in the freezer
(Ϫ23 °C) overnight. The cold liquid was decanted off and the crys-
tals were washed with cold DCE (ca. Ϫ20 °C, 2 ϫ 2.5 mL).
Alternatively, the crystals could be washed with a room tempera-
ture DCE/hexane mixture (1:2; 3 ϫ 2.5 mL). Subsequent recrystal-
lization from DCE (10 mL) afforded 4.92 g of product as white
crystals (92% yield). M.p. 133Ϫ136 °C. 13C NMR: δ ϭ 101.5 (tm,
JC,F ϭ 25.3 Hz), 118.7 (q, JC,F ϭ 317.8 Hz), 138.4 (dm, JC,F
ϭ
To estimate the relative activity of triflates 1 and 12 as
silylating agents we carried out a series of experiments. The
silylation reactions of camphor using either 1 or 12 and
triethylamine in CDCl3 were monitored by NMR spec-
troscopy.[21] In the experiment with TMSOTf 65% conver-
sion was observed in 12 h, while in the tube containing trifl-
ate 1 conversion into silyl enol ether was only 30% during
the same time. Qualitatively similar results were obtained
in experiments with methyl tert-butyl ketone, where with
TMSOTf more than 75% of the ketone was consumed by
the time of first acquisition (Ͻ1 min), whereas in the tube
with triflate 1 45% conversion was observed in five minutes.
These studies unambiguously demonstrate that, contrary
to our expectations, TPFS triflate 1 is a weaker silylating
agent than TMSOTf for the silylation of carbonyl com-
pounds. Such a phenomenon supports the conclusion that,
for the present system, the electronic effect of the penta-
fluorophenyl groups is overcompensated by steric factors.
258.9 Hz), 146.1 (dm, JC,F ϭ 263.1 Hz), 150.1 (dm, JC,F ϭ 248.4
Hz) ppm. 19F NMR: δ ϭ Ϫ158.82 (m), Ϫ142.86 (tt, JF,F ϭ 19.4,
5.6 Hz), Ϫ126.96 (dm, JF,F ϭ 16.8 Hz), Ϫ76.94 (s). C19F18O3SSi
(678.32): calcd. C 33.64; found C 33.58.
From TPFS Allyl: Triflic acid (0.31 mL, 3.57 mmol) was added to
a solution of TPFS allyl (0.98 g, 1.74 mmol) in DCM (1 mL). After
stirring for 20 min the solvent was evaporated, and the residue was
washed with hexane (3 ϫ 2 mL) and recrystallized from DCE
(2 mL) to give 0.87 g of 1 (73% yield).
From TPFS Enol Ether 3c: Triflic acid (0.62 mL, 6.97 mmol) was
added to a solution of 3c (3.71 g, 6.33 mmol) in DCM (5.5 mL).
After stirring for 5 min the solvent was evaporated, and the residue
recrystallized from DCE (8 mL) to give 3.72 g of 1 (87% yield).
Silyl Enol Ethers 3aϪk. General Procedure: Triethylamine
(0.83 mL, 6 mmol) and carbonyl compound 2aϪk (4 mmol) were
successively added, at 0 °C, to a suspension of triflate 1 (2.71 g,
4 mmol) in DCM (5.4 mL for 2aϪc,f, 2.7 mL for 2d,e,g-k). The
cooling bath was then removed and the mixture was stirred at room
temperature for the time indicated in Table 1. The solvent was eva-
porated, the residue was extracted with hexane (3 ϫ 5 mL) under
argon with the removal of the hexane phase through a cannula
covered with filter paper (for the extraction of 3a,f warm hexane,
ca. 50 °C, was used). The combined hexane solutions were concen-
Conclusion
We have described a new representative of the silyl triflate
family. TPFS triflate, which is readily accessible, serves as a trated, and the residue was distilled in vacuo.
Eur. J. Org. Chem. 2004, 5141Ϫ5148
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5145

A. D. Dilman et al.
FULL PAPER
[Tris(pentafluorophenyl)silyloxy]ethylene (3a): B.p. 122Ϫ124 °C/
was observed. C23H9F15OSi (614.38): calcd. C 44.96, H 1.48; found
0.08 Torr. M.p. 109Ϫ112 °C. 1H NMR: δ ϭ 4.43 (dd, J ϭ 5.6, C 44.91, H 1.50.
1.6 Hz, 1 H, CHAHBϭ), 4.70 (dd, J ϭ 13.5, 1.6 Hz, 1 H, CHAHBϭ
1-Phenyl-1-[tris(pentafluorophenyl)silyloxy]prop-1-ene (3h): E:Z ϭ
), 6.42 (dd, J ϭ 13.5, 5.6 Hz, 1 H, CHO) ppm. 13C NMR: δ ϭ
99.4, 104.2 (tm, JC,F ϭ 28.4 Hz), 137.8 (dm, JC,F ϭ 255.5 Hz),
142.9, 144.2 (dm, JC,F ϭ 258.3 Hz), 149.5 (dm, JC,F ϭ 246.9 Hz)
ppm. 19F NMR: δ ϭ Ϫ160.38 (m), Ϫ147.16 (t, JF,F ϭ 19.4 Hz),
Ϫ128.51 (dm, JF,F ϭ 19.4 Hz) ppm. C20H3F15OSi (572.30): calcd.
C 41.97, H 0.53; found C 42.15, H 0.55.
1
1:8.5. B.p. 140Ϫ142 °C/0.15 Torr. M.p. 103Ϫ111 °C. H NMR: Z-
isomer: δ ϭ 1.65 (d, J ϭ 6.9 Hz, 3 H, CH3), 5.31 (q, J ϭ 6.8 Hz, 1
H, CH), 7.12Ϫ7.33 (m, 5 H, Ph) ppm; E-isomer: δ ϭ 5.24 (q, J ϭ
7.5 Hz, 1 H, CH) ppm. 13C NMR: Z-isomer: δ ϭ 11.1, 105.0 (t,
JC,F ϭ 28.7 Hz), 108.3, 125.2, 128.0, 128.2, 137.46 (dm, JC,F
ϭ
257.6 Hz) 137.52, 143.9 (dm, JC,F ϭ 259.4 Hz), 148.5, 149.0 (dm,
1-[Tris(pentafluorophenyl)silyloxy]prop-1-ene (3b): E:Z ϭ 1:10. B.p.
JC,F ϭ 244.1 Hz) ppm; E-isomer: δ ϭ 107.5 ppm. 19F NMR: δ ϭ
1
120Ϫ122 °C/0.4 Torr. M.p. 95Ϫ100 °C. H NMR: Z-isomer: δ ϭ Ϫ160.63 (m), Ϫ147.32 (tt, JF,F ϭ 19.4, 5.5), Ϫ127.61 (dm, JF,F
ϭ
1.61 (dd, J ϭ 7.2, 1.6 Hz, 3 H, CH3), 4.86Ϫ4.75 (m, 1 H, CHCH3), 16.8 Hz) ppm. In the major isomer an NOE between the aromatic
6.19 (d, J ϭ 6.6 Hz, 1 H, CHO) ppm; E-isomer: δ ϭ 1.52 (dd, J ϭ protons and CH was observed. C26H9F15OSi (662.42): calcd. C
7.2, 1.7 Hz, CH3), 5.26 (dq, J ϭ 11.8, 7.2 Hz, 1 H, CHCH3) ppm. 48.96, H 1.37; found C 49.03, H 1.57.
13C NMR: Z-isomer: δ ϭ 9.1, 104.2 (tm, JC,F ϭ 26.9 Hz), 110.0,
1-[Tris(pentafluorophenyl)silyloxy]cyclohex-1-ene (3i): B.p. 122Ϫ125
135.8, 137.8 (dm, JC,F ϭ 256.7 Hz), 144.4 (dm, JC,F ϭ 258.5 Hz),
149.5 (dm, JC,F ϭ 247.7 Hz) ppm. 19F NMR: Z-isomer: δ ϭ
Ϫ160.37 (m), Ϫ146.76 (tt, JF,F ϭ 19.4, 5.6 Hz), Ϫ128.23 (dm,
JF,F ϭ 19.4 Hz) ppm. The configuration of the double bond was
determined by comparison of H-H coupling constants in the HCϭ
CH fragment (6.6 Hz in the Z-isomer vs. 11.8 Hz in the E-isomer).
C21H5F15OSi (586.33): calcd. C 43.02, H 0.86; found C 42.84, H
1.00.
°C/0.3 Torr. M.p. 71Ϫ72 °C. C24H9F15OSi (626.39): calcd. C 46.02,
H 1.45; found C 45.88, H 1.47. Spectral data are identical to those
reported earlier, except that the compound obtained here has none
of the impurities reported previously.[7]
1-[Tris(pentafluorophenyl)silyloxy]cyclohept-1-ene
(3j):
B.p.
144Ϫ146 °C/0.31 Torr. M.p. 62Ϫ64 °C 1H NMR: δ ϭ 1.39Ϫ1.74
[m, 6 H, (CH2)3], 1.94 (q, J ϭ 5.5 Hz, 2 H, CH2), 2.29Ϫ2.40 (m, 2
H, CH2), 5.02 (t, J ϭ 6.6 1 H, CH) ppm. 13C NMR: δ ϭ 25.0,
25.1, 27.5, 31.0, 34.5, 105.4 (tm, JC,F ϭ 28.4 Hz), 111.1, 137.7 (dm,
2-[Tris(pentafluorophenyl)silyloxy]propene (3c): Analytical and
spectroscopic data are identical to those reported previously.[7]
JC,F ϭ 255.5 Hz), 144.3 (dm, JC,F ϭ 258.3 Hz), 149.6 (dm, JC,F
ϭ
1-Phenyl-1-[tris(pentafluorophenyl)silyloxy]ethylene (3d): Viscous
248.4 Hz), 154.4 ppm. 19F NMR: δ ϭ Ϫ161.62 (m), Ϫ148.18 (t,
pale-yellow oil. B.p. 145Ϫ150 °C/0.4 Torr. C24H7F15OSi (648.39): JF,F ϭ 19.4 Hz), Ϫ128.30 (dm, JF,F ϭ 19.4 Hz) ppm.
calcld. C 48.16, H 1.09; found C 48.25, H 1.23. Spectral data are
1,7,7-Trimethyl-2-[tris(pentafluorophenyl)silyloxy]bicyclo[2.2.1]hept-
identical to those reported previously.[7]
2-ene (3k): B.p. 147Ϫ149 °C/0.08 Torr. M.p. 77Ϫ79 °C. [α]D20
ϭ
1
3,3-Dimethyl-2-[tris(pentafluorophenyl)silyloxy]but-1-ene (3e): B.p.
ϩ10.3 (c ϭ 3.60, CH2Cl2). H NMR: δ ϭ 0.72 (s, 3 H, CH3), 0.77
1
128Ϫ130 °C/0.45 Torr. M.p. 105Ϫ106 °C. H NMR: δ ϭ 1.08 (s, 9
(s, 3 H, CH3), 0.89 (m, 1 H), 0.94 (s, 3 H, CH3), 1.13 (m, 1 H),
H, tBu), 3.78 (d, J ϭ 3.0 Hz, 1 H, CHAHBϭ), 4.12 (d, J ϭ 3.0 Hz, 1.54 (m, 1 H), 1.82 (m, 1 H), 2.16 (t, J ϭ 3.3 Hz, 1 H, CH2CH2CH),
1 H, CHAHBϭ) ppm. 13C NMR: δ ϭ 27.8, 36.6, 88.7, 104.5 (tm,
JC,F ϭ 27.4 Hz), 137.6 (dm, JC,F ϭ 256.8 Hz), 144.2 (dm, JC,F
4.47 (d, J ϭ 3.3 Hz, 1 H, CHϭCO) ppm. 13C NMR (CDCl3): δ ϭ
9.6, 19.2, 20.0, 27.0, 30.9, 49.9, 53.7, 55.6, 104.5 (tm, JC,F ϭ 31.2
ϭ
258.9 Hz), 149.3 (dm, JC,F ϭ 246.3 Hz), 165.3 ppm. 19F NMR: δ ϭ Hz), 105.7, 137.7 (dm, JC,F ϭ 258.3 Hz), 144.4 (dm, JC,F ϭ 266.8
Ϫ160.45 (m), Ϫ147.10 (tt, JF,F ϭ 20.4, 4.5), Ϫ127.97 (dm, JF,F
ϭ
Hz), 149.5 (dm, JC,F ϭ 245.6 Hz), 158.0 ppm. 19F NMR: δ ϭ
20.4 Hz) ppm. C24H11F15OSi (628.41): calcd. C 45.87, H 1.76; Ϫ161.16 (m), Ϫ147.16 (tt, JF,F ϭ 19.4, 5.6 Hz), Ϫ128.01 (dm,
found C 46.06, H 2.02.
JF,F ϭ 19.4 Hz) ppm. C28H15F15OSi (680.48): calcd. C 49.42, H
2.22; found C 49.33, H 2.44.
1-Cyclopropyl-1-[tris(pentafluorophenyl)silyloxy]ethene (3f): B.p.
140Ϫ142 °C/0.5 Torr. M.p. 117Ϫ119 °C. H NMR: δ ϭ 0.56Ϫ0.70
1
Synthesis of 3l-term and 3m: A solution of triflate 1 (1.78 g,
(m, 4 H, CH2CH2), 1.41Ϫ1.54 (m, 1 H, CH), 3.70 (d, J ϭ 2.6 Hz,
2.62 mmol) in toluene (18 mL) was added with a cannula over
1 H, CHAHBϭ), 4.23 (d, J ϭ 2.6 Hz, 1 H, CHAHBϭ) ppm. 13C 20 min to a solution of ketone 2l or 2m (2.62 mmol) and triethyl-
NMR: δ ϭ 4.9, 15.1, 90.4, 104.4 (tm, JC,F ϭ 27.8 Hz), 137.7 (dm,
JC,F ϭ 254.9 Hz), 144.3 (dm, JC,F ϭ 260.3 Hz), 149.4 (dm, JC,F
252.2 Hz), 157.6 ppm. 19F NMR: δ ϭ Ϫ160.60 (m), Ϫ147.23 (tt,
JF,F 19.4, 5.5), Ϫ128.12 (dm, JF,F ϭ 19.4 Hz) ppm. the cooling bath. The solution phase was decanted and concen-
amine (0.55 mL, 3.93 mmol) in toluene (18 mL) at Ϫ78 °C. The
mixture was stirred for 6 hours at Ϫ78 °C, diluted with hexane
(18 mL), and allowed to warm to room temperature by removing
ϭ
ϭ
C24H11F15OSi (612.36): calcd. C 45.11, H 1.15; found C 45.23, H
1.22.
trated in vacuo. The residue was extracted twice with hexane (18
and 7 mL). The combined organic layers were concentrated to give
a crude product, which was either recrystallized or distilled.
3-[Tris(pentafluorophenyl)silyloxy]pent-2-ene (3g): E:Z ϭ 1:3.3. B.p.
1
120Ϫ122 °C/0.35 Torr. M.p. 36Ϫ45 °C. H NMR: Z-isomer: δ ϭ 3-Methyl-2-[tris(pentafluorophenyl)silyloxy]but-1-ene
(3lϪterm):
0.97 (t, J ϭ 7.2 Hz, 3 H, CH3CH2), 1.34 (d, J ϭ 6.9 Hz, 3 H, The crude product had a purity of about 95% which corresponds
CH3CH), 2.00 (q, J ϭ 7.2 Hz, 2 H, CH2), 4.62 (q, J ϭ 6.9 Hz, 1 to a yield of about 90%. As partial isomerization occurrs upon
H, CH) ppm; E-isomer: δ ϭ 1.01 (t, J ϭ 7.5 Hz, 3 H, CH3CH2),
1.46 (d, J ϭ 6.6 Hz, 3 H, CH3CH), 2.14 (q, J ϭ 7.5 Hz, 2 H, CH2),
distillation, the crude product was first recrystallized from hexane
(4 mL) and then distilled to give 1.33 g of 3l-term (63% yield). B.p.
1
4.49 (q, J ϭ 6.9 Hz, 1 H, CH) ppm. 13C NMR: Z-isomer: δ ϭ 11.1, 120Ϫ124 °C/0.03 Torr. M.p. 100Ϫ102 °C. H NMR: δ ϭ 1.05 (d,
23.6, 28.1, 103.1, 105.2 (m), 137.8 (dm, JC,F ϭ 254.0 Hz), 144.0 J ϭ 6.6 Hz, 6 H, 2CH3), 2.33 (sept, J ϭ 6.6 Hz, 1 H, CH), 3.83 (d,
(dm, JC,F ϭ 259.4 Hz), 149.4 (dm, JC,F ϭ 247.7 Hz), 150.9 ppm;
E-isomer: δ ϭ 10.3, 102.4, 151.9 ppm. 19F NMR: δ ϭ Ϫ160.59 (m), CHACHBϭ) ppm. 13C NMR: δ ϭ 20.2, 34.2, 89.9, 104.6 (tm,
Ϫ147.29 (tt, JF,F ϭ 19.4, 5.6), Ϫ127.98 (dm, JF,F ϭ 19.4 Hz) ppm. JC,F ϭ 26.3 Hz), 137.7 (dm, JC,F ϭ 256.7 Hz), 144.3 (dm, JC,F
In the major isomer an NOE between the CH2 and CH protons
260.3 Hz), 149.4 (dm, JC,F ϭ 247.7 Hz), 162.9 ppm. 19F NMR: δ ϭ
J ϭ 2.9 Hz, 1 H, CHACHBϭ), 4.14 (d, J ϭ 2.9 Hz, 1 H,
ϭ
5146
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 5141Ϫ5148

Synthesis and Silylation of Carbonyl Compounds
Ϫ160.84 (m), Ϫ147.51 (tt, JF,F ϭ 19.4, 5.5 Hz), Ϫ128.10 (dm, 19F NMR: δ ϭ Ϫ160.35 (m), Ϫ147.40 (tt, JF,F ϭ 20.4, 4.5 Hz)
FULL PAPER
JF,F ϭ 19.2 Hz) ppm. C24H9F15OSi (614.38): calcd. C 44.96, H
Ϫ127.34 (dm, JF,F ϭ 18.1 Hz) ppm. C24H9F15OSi (602.32): calcd.
1.48; found C 44.77, H 1.14.
C 41.88, H 0.84; found C 41.88, H 0.85.
1-Methoxy-1-[tris(pentafluorophenyl)silyloxy]ethene (6): A solution
of triflate 1 (0.687 g, 1.01 mmol) in methyl acetate (3.5 mL) was
added over 15 min to a solution of triethylamine (0.28 mL,
2.03 mmol) in methyl acetate (3.5 mL) at Ϫ78 °C. After stirring for
30 min at Ϫ78 °C the reaction flask was connected to a vacuum
pump and the low-temperature bath was replaced by an ice/water
bath. After concentration, the residue was extracted with hexane
at 0 °C (3 ϫ 5 mL). The combined extracts were concentrated in
vacuo at 0 °C to give 0.492 g of 6 as a colorless viscous oil (81%
Mixture of 2-[Tris(pentafluorophenyl)silyloxy]but-1-ene (3mϪterm)
and 2-[Tris(pentafluorophenyl)silyloxy]but-2-ene (3mϪint): The
crude product was distilled to give 1.88 g of a 1:1 mixture of ter-
minal and internal regioisomers in 81% yield; 3m-int is present as
a 1:10 mixture of E and Z regioisomers. B.p. 112Ϫ115 °C/0.05 Torr.
1
M.p. 83Ϫ85 °C. H NMR: 3m-term: δ ϭ 1.05 (t, J ϭ 7.9 Hz, 3 H,
CH3), 2.15 (q, J ϭ 7.9 Hz, 2 H, CH2CH3), 3.96 (d, J ϭ 2.6 Hz, 1
H, CHACHBϭ), 4.17 (d, J ϭ 2.6 Hz, 1 H, CHACHBϭ) ppm; (Z)-
3m-int: δ ϭ 1.37 (dq, J ϭ 6.6, 1.3 Hz, 3 H, CH3), 1.74 (s, 3 H,
CH3), 4.63 (q, J ϭ 6.6 Hz, 1 H, CH) ppm; (E)-3m-int: δ ϭ 1.47
(dq, J ϭ 7.2, 1.3 Hz, 3 H, CH3), 1.80 (s, 3 H, CH3) ppm. 13C NMR:
δ ϭ 10.2, 11.1, 21.2, 28.6, 91.5 (term), 105.7 (Z-int), 137.5 (dm,
JC,F ϭ 252.7 Hz), 144.0 (dm), 145.3, 149.2 (dm, JC,F ϭ 248.4 Hz),
159.1 ppm. 19F NMR: δ ϭ Ϫ160.70 (m), Ϫ147.34 (t, JF,F ϭ 19.4
Hz), Ϫ128.09 (m) ppm. In the major stereoisomer of 3m-int an
NOE between the CH3C(OSi) and CH protons was observed.
C26H9F15OSi (600.35): calcd. C 44.01, H 1.18; found C 43.88, H
1.39.
1
yield). H NMR: δ ϭ 3.29 (d, J ϭ 3.3 Hz, 1 H, CHACHBϭ), 3.49
(s, 3 H, CH3), 3.57 (d, J ϭ 3.3 Hz, 1 H, CHACHBϭ) ppm. 13C
NMR: δ ϭ 55.9, 62.7, 105.0 (td, JC,F ϭ 28.2, 3.1 Hz), 137.8 (dm,
JC,F ϭ 253.3 Hz), 144.3 (dm, JC,F ϭ 257.9 Hz), 149.6 (dm, JC,F
ϭ
247.2 Hz), 159.9 ppm. 19F NMR: δ ϭ Ϫ159.71 (m), Ϫ146.44 (t,
JF,F ϭ 19.4 Hz), Ϫ126.78 (dm, JF,F ϭ 19.4 Hz).
1-Methoxy-1-[tris(pentafluorophenyl)silyloxy]prop-1-ene (7): This
compound was obtained according to the general procedure using
5.4 mL of DCM for every 4 mmol of triflate 1; reaction time: 1.5 h.
Yield: 93%. Pale-yellow oil. B.p. 120Ϫ123 °C/0.3 Torr. 1H NMR:
δ ϭ 1.58 (d, J ϭ 6.6 Hz, 3 H, CH3CH), 3.37 (s, 3 H, OCH3), 3.66
(q, J ϭ 6.6 Hz, 1 H, CH) ppm. 13C NMR: δ ϭ 8.9, 55.4, 72.7,
105.3 (tm, JC,F ϭ 28.4 Hz), 137.6 (dm, JC,F ϭ 255.5 Hz), 144.1
(dm, JC,F ϭ 258.3 Hz), 149.4 (dm, JC,F ϭ 251.2 Hz), 154.7 ppm.
19F NMR: δ ϭ Ϫ161.25 (m), Ϫ148.18 (tt, JF,F ϭ 20.8, 5.6 Hz),
Ϫ128.56 (dm, JF,F ϭ 18.1 Hz) ppm. An NOE was observed be-
tween the OMe and CH protons.
2-[Tris(pentafluorophenyl)silyloxy]buta-1,3-diene (3n): Toluene
(2.2 mL), iPr2NEt (0.66 mL, 3.78 mmol), and 2n (256 µL,
3.15 mmol) were successively injected into a flask containing trifl-
ate 1 (2.16 g, 3.15 mmol) at room temperature. The mixture was
refluxed for 4 hours, then allowed to cool to 50 °C, diluted with
hexane (2.2 mL), and cooled to room temperature. The solution
phase was filtered under argon through a cannula covered with
filter paper, followed by washing the solid with hexane (2 ϫ 5 mL).
The combined hexane solution was concentrated, and the residue
was distilled in vacuo to afford 1.09 g of 3n as white crystals (58%
2-[Tris(pentafluorophenyl)silyloxy]-4,5-dihydrofuran (8): This com-
pound was obtained according to the general procedure using
5.4 mL of DCM for every 4 mmol of triflate 1; reaction time: 1.5 h.
Yield: 90%. The product decomposes upon distillation. M.p. 82Ϫ83
°C. 1H NMR: δ ϭ 2.62 (td, J ϭ 8.9, 2.0 Hz, 2 H, CH2CH2O), 3.96
(t, J ϭ 2.0 Hz, 1 H, CH), 4.22 (t, J ϭ 8.9 Hz, 2 H, CH2CH2O)
ppm. 13C NMR: δ ϭ 28.2, 69.3, 71.4, 104.1 (tm, JC,F ϭ 28.4 Hz),
137.2 (dm, JC,F ϭ 254.1 Hz), 144.2 (dm, JC,F ϭ 255.5 Hz), 149.3
(dm, JC,F ϭ 249.8 Hz), 155.5 ppm. 19F NMR: δ ϭ Ϫ161.39 (m),
Ϫ147.20 (tt, JF,F ϭ 19.4, 5.6 Hz), Ϫ128.15 (dm, JF,F ϭ 19.4 Hz)
ppm.
1
yield). B.p. 158Ϫ161 °C/0.55 Torr. M.p. 93Ϫ95 °C. H NMR: δ ϭ
4.25 (d, J ϭ 2.3 Hz, 1 H, CHACHBϭCO), 4.43 (d, J ϭ 2.3 Hz, 1 H,
CHACHBϭCO), 5.14 (d, J ϭ 10.7 Hz, 1 H), 5.51 (d, J ϭ 17.0 Hz, 1
H), 6.16 (dd, J ϭ 17.0, 10.5 Hz, 1 H) (CH2ϭCH) ppm. 13C NMR:
δ ϭ 97.7, 104.4 (tm, JC,F ϭ 27.0 Hz), 116.1, 132.4, 137.7 (dm,
JC,F ϭ 251.2 Hz), 144.3 (dm, JC,F ϭ 249.8 Hz), 149.5 (dm, JC,F
ϭ
249.8 Hz), 153.2 ppm. 19F NMR: δ ϭ Ϫ160.40 (m), Ϫ146.85 (tt,
JF,F ϭ 19.4, 5.6 Hz), Ϫ127.85 (dm, JF,F ϭ 19.4 Hz) ppm.
C22H5F15OSi (598.34): calcd. C 44.16, H 0.84; found C 43.83, H
0.90.
3-Methyl-2-[tris(pentafluorophenyl)silyloxy]-4,5-dihydrofuran
(9):
Ammonium Salt 4: CDCl3 (0.45 mL), triethylamine (35 µL,
0.25 mmol) and 2n (14.5 µL, 0.18 mmol) were successively added
to an NMR tube containing triflate 1 (115 mg, 0.17 mmol). The
tube was shaken and introduced into the NMR spectrometer. Salt
4 was formed as a 5:1 mixture of isomers. 1H NMR: major: δ ϭ
1.20Ϫ1.35 (br. t, 9 H, 3 CH3CH2), 1.95 (s, 3 H, CH3CϭC),
3.10Ϫ3.35 (br. q, 6 H, 3 CH3CH2), 3.81 (d, J ϭ 7.6 Hz, 2 H,
CH2CHϭC), 4.73 (t, J ϭ 7.5 Hz, 1 H, CH) ppm; minor: δ ϭ 1.84
(s, 3 H, CH3CϭC), 4.88 (t, J ϭ 7.5 Hz, 1 H, CH) ppm. 13C NMR:
major: δ ϭ 7.4, 17.6, 52.5, 55.3, 96.9, 156.9 ppm; minor: δ ϭ 7.1,
52.7, 96.7 ppm. 19F NMR: major: δ ϭ Ϫ159.87 (m), Ϫ145.86 (tt,
JF,F ϭ 19.4 Hz), Ϫ128.54 (d, JF,F ϭ 19.4 Hz), Ϫ79.90 (s) ppm;
minor: δ ϭ Ϫ159.31 (m), Ϫ145.18 (m) ppm.
This compound was obtained according to the general procedure
using 5.4 mL of DCM for every 4 mmol of triflate 1; reaction time:
1 h. The product decomposes upon distillation. Yield: 90%. M.p.
1
73Ϫ78 °C. H NMR: δ ϭ 1.56 (s, 3 H, CH3), 2.54 (t, J ϭ 8.5 Hz,
2 H, CH2CH2O), 4.08 (t, J ϭ 8.5 Hz, 2 H, CH2CH2O) ppm. 13C
NMR: δ ϭ 9.4, 33.1, 67.4, 79.8, 104.7 (tm, JC,F ϭ 27.8 Hz), 137.6
(dm, JC,F ϭ 255.5 Hz), 144.3 (dm, JC,F ϭ 259.8 Hz), 149.5 (dm,
J
C,F ϭ 248.4 Hz), 149.5 ppm. 19F NMR: δ ϭ Ϫ161.10 (m), Ϫ147.60
(tt, JF,F ϭ 19.4, 5.6 Hz), Ϫ128.23 (dm, JF,F ϭ 19.4 Hz) ppm.
X-ray Crystallographic Study: The diffraction data for 1 and 9 were
collected using Syntex P21 and Bruker Smart CCD 1000 dif-
˚
fractometers, respectively (Mo-Kα radiation, λ ϭ 0.71073 A). The
structures were solved by direct methods and refined by full-matrix
least-squares methods against F2 using the SHELXTL 5.1 pro-
gram.[24] The positions of the hydrogen atoms were calculated geo-
metrically and included in the refinement in a rigid-body approxi-
mation. The high thermal displacement parameters of the triflate
Methyl 2-Tris(pentafluorophenyl)silyl Acetate (5): This compound
was obtained according to the general procedure using 2.7 mL of
DCM for every 4 mmol of triflate 1; reaction time: 1 h. Yield: 90%.
B.p. 130Ϫ135 °C/0.35 Torr. M.p. 75Ϫ78 °C. 1H NMR: δ ϭ 2.98 (s,
2 H, CH2), 3.58 (s, 3 H, CH3) ppm. 13C NMR: δ ϭ 24.1, 52.5, group in the structure of 1 allowed us to split it into two positions
103.6 (tm, JC,F ϭ 28.5 Hz), 137.6 (dm, JC,F ϭ 254.7 Hz), 143.9
in a ratio of 7:3 and then refine them with the use of geometrical
(dm, JC,F ϭ 258.9 Hz), 149.4 (dm, JC,F ϭ 244.1 Hz), 169.7 ppm. restraints (SADI instruction).
Eur. J. Org. Chem. 2004, 5141Ϫ5148
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5147

A. D. Dilman et al.
FULL PAPER
of a carbocation at the CϭC double bond, which is very well
established, see: H. Mayr, B. Kempf, A. R. Ofial, Acc. Chem.
Res. 2003, 36, 66Ϫ77.
Crystal Data for 1: The crystal of C19F18O3SSi was triclinic at
¯
163 K, space group P1, a ϭ 10.158(2), b ϭ 11.228(2), c ϭ 11.284(2)
3
˚
˚
A, α ϭ 61.96(3)°, β ϭ 89.64(3)°, γ ϭ 74.84(3)°, V ϭ 1086.1(4) A ,
[9] [9a]
For the structure of Ph3SiOTf see: A. Asadi, A. G. Avent,
C. Eaborn, M. S. Hill, P. B. Hitchcock, M. M. Meehan, J. D.
Smith, Organometallics 2002, 21, 2183Ϫ2188. CCDC ref code
AGELED. [9b] According to the CCDC, the SiϪO bond length
Z ϭ 2, M ϭ 678.34, dcalc ϭ 2.074 g·cmϪ3, µ(Mo-Kα) ϭ 3.84 cmϪ1
,
F(000) ϭ 660. 5503 Reflections were measured, of which 5220 were
unique. R1 ϭ 0.0590 was calculated against F2 for 4523 reflections
with I Ͼ 2σ(I).
˚
in four-coordinate silyl triflates ranges from 1.742 to 1.760 A,
Crystal Data for 9: The crystal of C23H7F15O2Si was triclinic at
see structures AGELAZ, AGELED, KAVLAU, PODNEB,
XITFEL.
¯
120 K, space group P1, a ϭ 10.107(5), b ϭ 10.187(5), c ϭ 11.897(6)
[10]
˚
The problem of triflate passivation may become very important
if silylation is performed at temperatures at which the TPFS-
containing product may be poorly soluble.
A, α ϭ 80.255(10)°, β ϭ 81.481(11)°, γ ϭ 71.232(9)°, V ϭ
3
1137.3(10) A , Z ϭ 2, M ϭ 628.38, dcalc ϭ 1.835 g·cmϪ3, µ(Mo-
˚
Kα) ϭ 2.49 cmϪ1, F(000) ϭ 620. 6249 Reflections were measured,
of which 4292 were unique. R1 ϭ 0.0686 was calculated against F2
for 2832 reflections with I Ͼ 2σ(I).
[11] [11a] P. Cazeau, F. Duboudin, F. Moulines, O. Babot, J. Dunog-
ues, Tetrahedron 1987, 43, 2075Ϫ2088. [11b] H. Emde, A. Götz,
K. Hofmann, G. Simchen, Liebigs Ann. Chem. 1981,
1643Ϫ1657. [11c] For a pertinent discussion see: S. E. Denmark,
S. M. Pham, J. Org. Chem. 2003, 68, 5045Ϫ5055.
CCDC-245733 (for 1) and -245734 (for 9) contain the supplemen-
tary crystallographic data for this paper. These data can be ob-
tained free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html
[or from the Cambridge Crystallographic Data Centre, 12, Union
Road, Cambridge CB2 1EZ, UK; fax: ϩ44-1223-336-033; E-mail:
deposit@ccdc.cam.ac.uk].
[12]
Formation of salts similar to 4 has been reported in the litera-
ture, see ref. 11a.
Vacuum distillation of 3 usually posed no problems. However,
[13]
compound 3l-term, produced as a single regioisomer, under-
went partial isomerization into 3l-int upon distillation, presum-
ably due to the presence of traces of triethylammonium triflate.
Therefore, 3l-term had to be recrystallized before distillation.
[14] [14a] Rearrangement of trichlorosilyl ketene acetals to α-trichlo-
silyl esters has been reported, see: I. F. Lutsenko, Yu. I. Baukov,
Supporting Information (see also the footnote on the first page of
1
this article): H, 13C, and 19F NMR spectra of 3jϪl and 5Ϫ9.
G. S. Burlachenko, B. N. Khasapov, J. Organomet. Chem. 1966,
Acknowledgments
[14b]
5, 20Ϫ28.
The structural assignment between isomers 5
and 6 was unequivocally made by 13C NMR spectroscopy,
where the chemical shifts of the α-carbon atom differ by
39 ppm (sp3 vs. sp2), see: G. L. Larson, V. G. de Maldonado,
L. M. Fuentes, L. E. Torres, J. Org. Chem. 1988, 53, 633Ϫ639.
The configuration was determined by an NOE experiment.
In CDCl3 solution no reaction was observed at room tempera-
ture after 5 h. After 48 h 25% of methyl isobutyrate had disap-
peared, although not to give the silyl ketene acetal or α-silyl
ester, but to some other species, which were tentatively ident-
ified as TPFS isobutyrate and methyltriethylammonium trifl-
ate.
This work was performed at the Scientific Educational Center for
Young Chemists and was supported by the Ministry of Science
(project MK-2352.2003.03), the Russian Foundation for Basic Re-
search (project no. 00-15-97359), and INTAS (project no. 2003-55-
1185). We also thank Prof. S. L. Ioffe for start-up funding and Dr.
V. M. Danilenko for helpful discussions.
[15]
[16]
[1] [1a]
T. W. Greene, P. G. M. Wuts, Protective Groups in Organic
Synthesis, John Wiley & Sons, New York, 1999. [1b] P. J. Kocien-
ski, Protecting Groups, George Thieme Verlag, Stuttgart, 1994.
[2] [2a]
[17]
A. D. Dilman, S. L. Ioffe, Chem. Rev. 2003, 103, 733Ϫ772.
Only two structures of silyl ketene acetals were found in the
[17a]
[2b] M. Oishi, in Lewis Acids in Organic Synthesis (Ed.: H. Yam-
amoto), Wiley-VCH, Weinheim, 2000, vol. 1, pp. 355Ϫ393.
For recent advances see:
Cambridge database:
(E)-tBuPh2SiO(tBuO)CϭCHMe,
˚
SiϪO 1.660 A, CCDC ref code VAKYUB; R. E. Babston, V.
Lynch, C. S. Wilcox, Tetrahedron Lett. 1989, 30, 447Ϫ450. [17b]
1,1,2,2-Tetrakis(2,4,6-triisopropylphenyl)-1,2-dihydrofuro[3,2-
[3]
[3a]
X ϭ C6F5CTf2: A. Hasegawa, K.
Ishihara, H. Yamamoto, Angew. Chem. Int. Ed. 2003, 42,
5731Ϫ5733. [3b] X ϭ NTf2: B. Mathieu, L. Ghosez, Tetrahedron
d][1,2,3]oxadisilol-5(6H)-one, SiϪO 1.725 A, CCDC ref code
˚
[3c]
2002, 58, 8219Ϫ8226.
X ϭ N(SO2F)2: G. Kaur, A. Trehan,
QOXDIQ: S. Boomgaarden, W. Saak, M. Weidenbruch, H.
Marsmann, Organometallics 2001, 20, 2451Ϫ2453.
H. Emde, G. Simchen, Synthesis 1977, 867Ϫ869.
Cations similar to 10, containing a trialkylsilyl fragment, have
been previously identified by NMR spectroscopy when using a
non-coordinating counterion, see: G. K. S. Prakash, Q. Wang,
G. Rasul, G. A. Olah, J. Organomet. Chem. 1998, 550,
119Ϫ123.
A. L. Klebanskii, Yu. A. Yuzhelevskii, E. G. Kagan, G. A.
Nikolaev, A. V. Harlamova, Zh. Obsch. Khim. 1968, 38,
914Ϫ918 (Russ. J. Gen. Chem. 1968, 38, 878Ϫ882).
For quantitative NMR studies a homogeneous solution of
TPFS triflate 1 was used at a concentration of about 50 mg of
1 per 0.5 mL of CDCl3.
A. D. Dilman, D. E. Arkhipov, A. A. Korlyukov, V. P. Anani-
kov, V. M. Danilenko, V. A. Tartakovsky, unpublished work.
F. N. Stepanov, Zh. Obshch. Khim. 1955, 25, 2480Ϫ2485
(Chem. Abstr. 1956, 50:9291f).
G. M. Sheldrick, SHELXTL-97, V5.10, 1997, Bruker AXS
Inc., Madison, WI-53719, USA.
S. Trehan, J. Org. Chem. 1998, 63, 2365Ϫ2366.
H. Emde, D. Domsch, H. Feger, U. Frick, A. Gotz, H. H.
Hergott, K. Hofmann, W. Kober, K. Krageloh, T. Oesterle, W.
[4] [4a]
[18]
[19]
[4b]
Steppan, W. West, G. Simchen, Synthesis 1982, 1Ϫ26.
G.
Simchen, in Advances in Silicon Chemistry (Ed.: G. L. Larson),
JAI Press, Inc., Greenwich, CT, 1991, vol. 1, pp 189Ϫ301. [4c]S.
Murata, M. Suzuki, R. Noyori, Tetrahedron 1988, 44,
4259Ϫ4275.
[20]
[21]
[5]
The pentafluorophenyl group is frequently used to modify the
[5a]
properties of boron-based Lewis acids.
T. Schwier, M. Ru-
bin, V. Gevorgyan, Org. Lett. 2004, 6, 1999Ϫ2001. [5b] K. Ishih-
ara, H. Yamamoto, Eur. J. Org. Chem. 1999, 527Ϫ538.
[5c]
J.
M. Blackwell, W. E. Piers, R. McDonald, J. Am. Chem. Soc.
2002, 124, 1295Ϫ1306.
[22]
[23]
[24]
[6]
[7]
H.-J. Frohn, A. Lewin, V. V. Bardin, J. Organomet. Chem. 1998,
568, 233Ϫ240.
A. D. Dilman, P. A. Belyakov, A. A. Korlyukov, V. A. Tar-
takovsky, Tetrahedron Lett. 2004, 45, 3741Ϫ3744.
[8] [8a]
[8b]
W. Uhlig, Chem. Ber. 1996, 129, 733Ϫ739.
The rate of
protodesilylation is thought to correlate with the rate of attack
Received August 8, 2004
5148
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 5141Ϫ5148