Crucial role of the conjugate base for silyl Lewis acid induced Mukaiyama aldol reactions

Article abstract of DOI:10.1002/ejoc.200500845

The silyl Lewis acid induced Mukaiyama aldol reaction proceeds through each catalytic cycle under the influence of their conjugate bases; there is an especially significant difference between the low nucleophilic conjugate bases, ^-NTf₂ and ^-CTf₃, and the relatively high nucleophilic ^-OTf. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejoc.200500845

FULL PAPER
DOI: 10.1002/ejoc.200500845
Crucial Role of the Conjugate Base for Silyl Lewis Acid Induced Mukaiyama
Aldol Reactions
[a]
[b]
[a]
Yukihiro Hiraiwa, Kazuaki Ishihara, and Hisashi Yamamoto*
Keywords: Mukaiyama aldol reaction / Reaction mechanism / Silyl Lewis acid / Conjugate base
The silyl Lewis acid induced Mukaiyama aldol reaction pro- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
ceeds through each catalytic cycle under the influence of
Germany, 2006)
their conjugate bases; there is an especially significant differ-
–
ence between the low nucleophilic conjugate bases, NTf
and CTf , and the relatively high nucleophilic OTf.
3
2
–
–
Introduction
Mechanistic investigations for Lewis acid induced Mu-
kaiyama aldol reactions have been demonstrated over the
[
1–4]
last decade.
The possible and proposed pathways for the
Lewis acid catalyzed Mukaiyama aldol reaction of silyl enol
ether 2 with the aldehyde 1 are shown in Scheme 1. The
reaction of the activated aldehyde 4 and the silyl enol ether
[
5]
2
generates the silyloxycarbenium intermediate 5a or 5b.
The intermolecular transfer of the silyl group to conjugate
base X produces the metal aldolate 6 and the silyl Lewis
3
acid (R SiX) derived from silyl enol ether (path A). This
3
reaction pathway may account for the relatively high nu-
cleophilicity of ligand X. When the metathesis between 6
3
3
and R SiX is faster than the coordination of 1 to R SiX,
3
3
the catalyst MX regenerates to give the silyl aldolate 3 (path
C). On the other hand, the silyl Lewis acid catalysis eventu-
ates from the slow metathesis (path D). Meanwhile, at the
stage of intermediate 5a or 5b, the silyl group can be cap-
tured by the carbonyl substrate (path B). In this case, the
Lewis acid catalyst regenerates directly. Among these pos-
sible reaction mechanisms, however, the common conclu-
sions in notable mechanistic studies can be summarized as
follows: (1) the generation of the silyl Lewis acid derived
from silyl enol ether, and (2) one of the real catalysts is
Scheme 1.
derived from this powerful silyl Lewis acid generated in situ
[
6–9]
rather than from the Lewis acid catalyst.[2–4]
Recently, we demonstrated that Me
3
SiNTf
2
shows a
strong catalytic activity for the Mukaiyama aldol reaction
[
10]
and the Sakurai-Hosomi allylation reaction.
The conju-
–
gate base, NTf , is well recognized as a strong electron-
2
withdrawing and weakly coordinating counteranion, that is
[
11]
[
[
a] The Department of Chemistry, The University of Chicago,
a ligand of low nucleophilicity. The counteranion of tris-
5
735 South Ellis Avenue, Chicago, Illinois 60637, USA
–
(
triflyl)methane ( CTf ) is also known as a weakly coordi-
3
Fax: +1-773-702-0805
E-mail: yamamoto@uchicago.edu
b] Graduate School of Engineering, Nagoya University,
Furo-cho, Chikusa, Nagoya, 464-8603, Japan
Fax: +81-52-789-3222
nating counteranion, and the corresponding Brønsted acid
(
HCTf ) was reported to be one of the strongest acids in
3
[
12–14]
the gas phase.
To date, the mechanistic studies on
E-mail: ishihara@cc.nagoya-u.ac.jp
Lewis acid induced Mukaiyama aldol reactions reported
Eur. J. Org. Chem. 2006, 1837–1844
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1837

Y. Hiraiwa, K. Ishihara, H. Yamamoto
FULL PAPER
have lacked any focus on the role of conjugate base on the silyl aldolates 10 and 11 when the reaction is conducted in
Lewis acid catalyst. Herein, we describe the diversity of the the presence of 1.0 equiv. of Me SiX.
3
reaction mechanism, which is dependent on the conjugate
base in silyl Lewis acid induced Mukaiyama aldol reac-
[
15]
tions.
Mechanistic Study on the R SiNTf -Induced Mukaiyama
3
2
Aldol Reaction
Firstly, the mechanistic details of the Me SiNTf -in-
duced Mukaiyama aldol reaction were investigated. To clar-
Results and Discussion
3
2
Initially, we considered two extreme mechanisms for silyl ify whether the transformation from 12a or 12b to a silyl
Lewis acid induced Mukaiyama aldol reactions.[4] These are aldolate occurs through X (path A) or R Si transfer (path
–
+
3
illustrated in Scheme 2, where the Mukaiyama aldol reac- B), the aldol reaction of tBuMe Si enol ether 9 with benzal-
2
tion of tBuMe Si-enol ether 9 with benzaldehyde (8) is an dehyde (8) was carried out in the presence of 1.0 equiv. Me3-
2
example and where a trimethylsilyl Lewis acid (Me SiX) is SiNTf (Scheme 3). To control the strong Lewis acidity of
3
2
employed as the catalyst. In the first stage of the catalytic Me SiNTf , Et O was used as a solvent for the experi-
3
2
2
[
10]
process, the addition of the silyl enol ether 9 to the aldehyde ments. Regardless of the addition of substrates and Me3-
should generate a siloxocarbenium ion intermediate 12a SiNTf , only the tBuMe Si aldolate 11 was produced. In
8
2
2
or 12b. In path A, tBuMe SiX is released from the interme- contrast, the reaction of Me Si enol ether 13 with 8 in the
2
3
diate 12a or 12b to give the Me Si aldolate 10 by intramol- presence of 1.0 equiv. of tBuMe SiNTf gave the exclusive
3
2
2
–
ecular transfer of the counteranion X . Thus, tBuMe SiX formation of Me Si aldolate 10 in moderate yield. Further-
2
3
emerges as the catalyst for the next reaction path. On the more, neither the silyl group of tBuMe Si aldolate 11 and
2
other hand, in path B, the Me SiX is released into the reac- Me Si aldolate 10 nor that of silyl enol ethers 9 and 13
3
3
[
16]
tion medium to provide the tBuMe Si aldolate 11 by intra- was exchanged under similar reaction conditions.
These
2
+
molecular transfer of tBuMe Si . Therefore, Me SiX plays experimental results would exclude the silyl catalysis path
2
3
a role as a catalyst in path B. We assumed that this reaction A derived from silyl enol ether, and indicate the possibility
mechanism can be expected from the ratio of the resulting of path B.
Scheme 2.
Scheme 3.
838
1
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1837–1844

Role of the Conjugate Base for Silyl Lewis Acid Induced Mukaiyama Aldol Reactions
FULL PAPER
However, the double-label crossover experiment, in a similar conditions. These facts unambiguously preclude
manner similar to that described by Carreira and Den- path B and suggest the silyl group transfer between interme-
[
2,3]
mark,
demonstrated the feasibility of path A (Scheme 4). diates.
While the intermolecular silyl group transfer is in conflict
Two silyl enol ethers of similar steric and electronic de-
mands (9 and 14) were prepared and combined with benzal- with the afore-mentioned results for the control experi-
dehyde (8) in the presence of 1 mol-% of Me SiNTf . The ments in Scheme 3, the investigations of the syn/anti selec-
3
2
reaction was complete at –78 °C within 15 min, and the tivities in R SiNTf -induced aldol reactions advocates the
3
2
product composition was determined by GC analysis. A exchange of the silyl group in the reaction course. The reac-
mixture of scrambled silyl aldolates was isolated in a ratio tions of (E)- and (Z)-silyl enol ethers 18–21 with aldehyde
of 1:1.0:0.89:0.81 (11/15/16/17). Control experiments 8 were conducted in the presence of 3 mol-% of R SiNTf2
3
showed that neither the starting silyl enol ethers 9 and 14 (Table 1). Similar syn/anti ratios of the products were ob-
nor the silyl aldolates 11 and 15 were exchanged under- tained, which are independent of the size of the trialkylsilyl
group of the catalyst. In contrast, the diastereoselectivities
correlate to the size of the silyl groups of the silyl enol ether.
These experimental data indicate that the catalysis is initi-
ated by the silyl group of silyl enol ether.
Plausible Mechanism for the R SiNTf -Induced
3
2
Mukaiyama Aldol Reaction
In order to explain the results of the control experiments
Scheme 3) and the crossover experiment (Scheme 4), it was
(
imperative to postulate another mechanism for the
R SiNTf -induced Mukaiyama aldol reaction, since neither
3
2
path A nor B could account for the results. With the low
–
nucleophilicity of NTf in mind, we proposed the novel
2
reaction mechanism E for the R SiNTf -induced Mukai-
3
2
yama aldol reaction (Scheme 5). In path Ea, the direct in-
Scheme 4.
3
terception of the silyl group (–SiR ) in intermediate 23a or
3
2
3b by aldehyde 1 would provide the silyloxycarbenium cat-
ion intermediate 24. Nucleophilic attack of silyl enol ether
on 24 would generate the other silyloxycarbenium cation
2
Table 1. syn/anti Selectivity in the Mukaiyama aldol reaction.
intermediate 26, and the reaction would proceed in the
same manner. Additionally, we considered the role of Et O
2
for the alternative reaction mechanism (path Eb). Since
Et O can coordinate to a silyl Lewis acid, a highly reactive
2
3
+
– [17]
silyloxycarbenium cation, R Si–OEt [counteranion] ,
3
2
may be generated in the reaction medium, especially with
–
the low nucleophilic counteranion NTf .
2
In this reaction pathway E, two possible counteranions
could be possible. One is the silicate anion 25, which bears
–
an NTf group on the silyl aldolate, the other is NTf ,
2
2
which is released from the silyl Lewis acid. In both cases, it
is difficult to distinguish path E from path D. However, we
envisaged that reaction path D could not reasonably explain
the results of the control experiments in Scheme 3. The re-
action in the presence of 1.0 equiv. of silyl triflylimide
should produce a definite amount of silyl aldolate by path
D, which bears the silyl group of silyl triflylimide. The real
activation species in path E would be the silyloxycarbenium
cation intermediate 24 or 26, and either would be expected
to be a stronger Lewis acid than Me SiNTf and tBuMe S-
1
[
(
a] (E)/(Z) = 1:99, determined by H NMR spectroscopy. [b] (E)/
3
2
2
1
Z) = 4:96, determined by H NMR spectroscopy. [c] Prepared in
situ by the treatment of AgNTf with the corresponding silyl chlo-
ride in Et O at room temp. for 0.5 h. [d] Determined by H NMR
[
17]
iNTf2.
Our proposed mechanism E and our hypothesis
2
1
satisfactorily interprets all the control experiments and the
double-label crossover experiments.
2
spectroscopy.
Eur. J. Org. Chem. 2006, 1837–1844
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1839

Y. Hiraiwa, K. Ishihara, H. Yamamoto
FULL PAPER
Scheme 5.
Mechanistic Study on the R SiCTf -Induced Mukaiyama
of the reaction of tBuMe Si enol ether 9 with benzaldehyde
3
3
2
Aldol Reaction
(8) in the presence of 1.0 equiv. of Me SiCTf , tBuMe Si
3
3
2
aldolate 11 was produced exclusively in 63% yield. No
1
Next, the mechanistic study of the R SiCTf -induced Me Si aldolate was detected by H NMR spectroscopy or
3
3
3
Mukaiyama aldol reaction was inspected. Me SiCTf and MS analysis. In contrast, the use of 1.0 equiv. of tBuMe S-
3
3
2
tBuMe SiCTf were prepared in situ by the treatment of iCTf for the reaction of Me Si enol ether 13 with benzalde-
2
3
3
3
[18]
Me SiCl and tBuMe SiCl with AgCTf
in Et O at room hyde (8) produced Me Si aldolate 10 in 34% yield.
3
2
3
2
3
temperature, respectively, and were used without further
purifications. The control experiments were conducted in a
A double-label crossover experiment was also conducted
for the Me SiCTf -induced Mukaiyama aldol reaction un-
der the same reaction conditions as those employed for Me3-
3
3
manner similar to that for R SiNTf (Scheme 6). In the case
3
2
Scheme 6.
840
1
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1837–1844

Role of the Conjugate Base for Silyl Lewis Acid Induced Mukaiyama Aldol Reactions
FULL PAPER
SiNTf (Scheme 7). Four silyl aldolates (11, 15–17) were the intermediate 25. The same trend of products in the
2
produced in quantitative yield, and the ratio was similar to R SiCTf - as in the R SiNTf -induced Mukaiyama aldol
3
3
3
2
the results of Me SiNTf . The small amount of desilylated reaction led us to the conclusion that the former also pro-
3
2
–
aldol 27 was confirmed by GC analysis and may indicate ceeds through path E. The low nucleophilicity of CTf akin
3
to that of –NTf2 probably contributes to the reaction
mechanism.
Mechanistic Study on the R SiOTf-Induced Mukaiyama
3
Aldol Reaction
Finally, we performed a mechanistic study of the R Si-
3
[
4,19]
OTf-induced Mukaiyama aldol reaction.
The control
reactions analogous to those of Scheme 3 were conducted
using Me SiOTf (Table 2). When the reaction was carried
3
out at –100 °C for 15 min, Me Si aldolate 10 was obtained
3
exclusively in 24% yield, along with the starting materials
(
Table 2, Entry 1). With the Me SiNTf - or Me SiCTf -in-
3 2 3 3
duced reaction, the ratio of silyl aldolates 10 and 11 was
completely reversed with respect to that obtained with Me3-
SiOTf. The ratio of produced silyl aldolates suggests that
Scheme 7.
3
Table 2. Control experiment for Me SiOTf-induced Mukaiyama aldol reaction.
Entry
Conditions
Isolated yield [%]
Ratio 10/11[a]
1
2
–100 °C, 0.5 h, 0.02 
–78 °C, 5 h, 0.06 
24
61
Ͼ99:1
83:17
1
[
a] Determined by H NMR spectroscopy for the mixture of 10 and 11.
Scheme 8.
Eur. J. Org. Chem. 2006, 1837–1844
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1841

Y. Hiraiwa, K. Ishihara, H. Yamamoto
FULL PAPER
the Me SiOTf-induced Mukaiyama aldol reaction proceeds (125 MHz) and 400 (100 MHz), and Varian Gemini-300 (75 MHz)
3
spectrometers at room temperature. Chemical shifts are reported in
ppm from the solvent resonance employed as the internal standard
through path A. The starting materials are not completely
consumed upon prolonged stirring at –78 °C even with a
(
CDCl
raphy (GC) was performed with a Shimadzu Model GC-17A Ver.
instrument with a flame-ionization detector and a capillary col-
3
at δ = 77.00 ppm). Analytical gas–liquid phase chromatog-
concentration threefold higher than that in the reaction at
100 °C, and the mixture of 10 and 11 was obtained in 61%
–
3
yield at a molar ratio of 83:17 (Table 2, Entry 2).
umn of TC-1 (100% dimethylpolysiloxane, 0.25 mm×30 m, GL
Sciences Inc.) using helium as carrier gas (116.0 kPa). Measure-
ment conditions: Injection temp. 150 °C; column temp. 70 °C for
It seems that the result of Entry 2 does not match path A
towing to the generation of tBuMe Si aldolate 11. However,
2
further control experiments offered an idea to explain the 5 min, +10 °C/min for 18 min, and 250 °C for 7 min. High-resolu-
reaction mechanism adequately by path A. For instance, tion mass spectra data were obtained from the Research Resources
tBuMe SiOTf cannot promote the reaction of tBuMe Si Center Mass Spectrometry Laboratory at University of Illinois at
2
2
Chicago. For thin-layer chromatography (TLC) analysis through-
out this work, Merck precoated TLC plates (silica gel 60 GF254
enol ether with benzaldehyde (8) under similar reaction
conditions as those in Table 2. Furthermore, control experi-
ments of the scrambling for the silyl groups among silyl
0
.25 mm) were used. The products were purified by flash column
chromatography on silica gel (E. Merck 9325). In experiments that
required dry solvents, CH Cl , Et O and THF were stored under
dry argon after being distilled from calcium hydride (CH Cl ) or
sodium metal using benzophenone ketyl as an initiator (Et O and
THF), and were then used. Simple chemicals (if not described in
aldolates 11, silyl enol ether 9 and Me SiOTf showed that
3
2
2
2
neither 9 and Me SiOTf nor 11 and Me SiOTf were ex-
3
3
2
2
changed under similar conditions. On the basis of our ob-
2
servations and Bosnich’s proposed pathway,[ the mecha-
4]
nism for the R SiOTf-induced Mukaiyama aldol reaction this Exp. Sect.) were purchased and used. AgNTf and AgCTf3
3
2
could be concluded to be an intramolecular transfer of were prepared according to Shreeve’s procedure.[
OTf (path A), and the tBuMe SiOTf catalysis is shown in
Scheme 8 as a model case. From intermediate 28a or 28b,
18,21]
–
[6]
2
3 2 3 2
Preparation of Me SiNTf : Me SiNTf was prepared from freshly
distilled Me SiCl (1.2 equiv.) and AgNTf (1.0 equiv.) in CH Cl
3 2 2 2
Me SiOTf is released into the reaction medium by the intra- and purified by distillation (80–84 °C, 7 Torr). Me SiNTf was also
3
3
2
–
molecular transfer of OTf, and the reaction is promoted prepared from freshly distilled Me
in the same manner. Since Me SiOTf possesses a stronger (1 equiv.) in Et
reactivity than tBuMe SiOTf, the Me Si aldolate 10 would
3
SiCl (1.2 equiv.) and AgNTf
O and used for the reaction without further purifi-
2
2
3
cation.
2
3
be generated through Me SiOTf catalysis under the reac- Preparation of tBuMe SiNTf , Me SiCTf , and tBuMe SiCTf :
3
2
2
3
3
2
3
tion conditions shown in Table 2. In the case of the R Si- These silyl Lewis acids were prepared in situ according to the pro-
3
cedure for Me
silver salt.
3 2
SiNTf using the corresponding silyl chloride and
OTf-induced Mukaiyama aldol reaction, the formation of
+
–
Me Si·OEt with OTf can be excluded because of the rel-
3
2
–
[20]
atively high nucleophilicity of OTf.
General Procedure for the Control Experiments (Schemes 3 and 6,
Table 2): To a solution of in situ prepared tBuMe SiNTf
1.0 mmol) in dry Et O (50 mL) was added freshly distilled benzal-
dehyde (8, 102 µL, 1.0 mmol) at –100 °C. Subsequently, Me Si enol
ether 13 (281 mg, 1.2 mmol) was added in a dropwise fashion at
2
2
(
2
Conclusion
3
We have demonstrated the diversity of the mechanism for the same temperature. After stirring for 0.5 h, pyridine (200 µL)
silyl Lewis acid induced Mukaiyama aldol reactions that and saturated aqueous NaHCO (20 mL) were added. The mixture
are dependent on the conjugate base, and have proposed a was warmed to room temperature and extracted with Et
2×20 mL). The combined organic extracts were dried with
Na SO , concentrated under reduced pressure, and the residue was
purified by flash column chromatography on silica gel (hexane/
EtOAc, 10:1) to give Me Si aldolate 10 (185 mg, 62% yield). The
corresponding tBuMe Si aldolate 11 was not formed in a detectable
3
2
O
(
novel reaction mechanism E. In the case of R SiNTf - or
3
2
2
4
R SiCTf -induced Mukaiyama aldol reactions, the low nu-
3
3
cleophilicity of their conjugate bases prevent not only the
intramolecular transfer of X (path A) but also that of the
–
3
2
R Si group to the carbonyl oxygen atom (path B). On the
1
3
amount ( H NMR and GC analysis).
–
[17]
other hand, the high nucleophilicity of OTf
intramolecular transfer of X and generates the silyl Lewis
acid derived from silyl enol ether.
causes the
–
General Procedure for the Silyl Group Scrambling Control Experi-
ments (Schemes 3, 4, 6 and 7, Table 2): To a solution of 3-(tert-
butyldimethylsilyloxy)-1,3-diphenylpropan-1-one (11, 340 mg,
1
(
0
.0 mmol) in Et
184 µL, 1.0 mmol) at –100 °C. The reaction mixture was stirred for
.5 h, and quenched with pyridine (200 µL) and saturated aqueous
NaHCO (20 mL). The mixture was then warmed to room tem-
perature and extracted with Et O (2×20 mL). The combined or-
ganic extracts were dried with Na SO , and concentrated under
reduced pressure. The corresponding Me Si aldolate 10 was not
formed in a detectable amount ( H NMR and GC analysis).
2 3
O (50 mL) was added freshly distilled Me SiOTf
Experimental Section
3
General Remarks: Unless otherwise stated, reactions were per-
formed in flame-dried glassware under argon. Infrared (IR) spectra
were recorded with a Shimadzu FTIR-8100 spectrometer and Nico-
let 20 SXB FTIR spectrometer, and are reported in reciprocal centi-
2
2
4
3
1
–1
1
meters (cm ). H NMR spectra were measured with Bruker DMX-
00 (500 MHz), 400 (400 MHz), and Varian Gemini-300 Procedure for the Silyl Group Scrambling Control Experiments of
300 MHz) spectrometers at room temperature. Data are reported Silyl Enol Ether 9 and 14 (Schemes 4 and 7): To a solution of 9
(1.8 mmol, 422 mg), 14 (1.8 mmol, 498 mg), and Me SiNTf
(0.03 mmol, prepared in situ by the treatment of Me SiCl and
AgNTf ) in Et O (6.0 mL) was added benzaldehyde (0.3 mmol,
5
(
as follows: chemical shift in ppm from internal Me
4
Si (δ = 0.0 ppm)
3
2
on the δ scale, coupling constant J [Hz], integration, and assign-
3
ment. 13C NMR spectra were recorded with Bruker DMX-500
2
2
1842
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1837–1844

Role of the Conjugate Base for Silyl Lewis Acid Induced Mukaiyama Aldol Reactions
FULL PAPER
3
1
0 µL) at –78 °C. The reaction mixture was stirred at –78 °C for
5 min and quenched with pyridine (5 drops) followed by saturated
2
silyl enol ether 14 (166 mg, 0.6 mmol) in Et O (1 mL) was added
freshly distilled benzaldehyde (8, 51 µL, 0.5 mmol) at –78 °C. The
aqueous NaHCO
perature and extracted with EtOAc. The combined organic layers
were dried with Na SO and concentrated. No silyl group exchange
3
(15 mL). The mixture was warmed to room tem-
reaction mixture was stirred for 15 min, quenched with pyridine (2
drops) and a saturated aqueous NaHCO
warmed to room temperature, and extracted with Et
The combined organic layers were dried with Na SO
4
3
solution (5 mL), then
O (3×50 mL).
, concentrated
2
4
2
1
was observed by H NMR analysis of the residue.
2
in vacuo, and purified by flash column chromatography on silica
gel (hexane/EtOAc, 50:1) to give 15 (195 mg, 99% yield) as a color-
General Procedure for the Double-Label Crossover Experiment
(
0
Schemes 4 and 7): To
.03 mmol), silyl enol ether 9 (421 mg, 1.8 mmol), and silyl enol
ether 14 (497 mg, 1.8 mmol) in Et O (6 mL) was added freshly dis-
a
solution of Me
3
SiNTf
2
(11 mg,
f R
less oil. TLC (hexane/EtOAc, 4:1): R = 0.64. GC: t = 22.8 min.
IR (film): ν˜ = 2959, 1684, 1607, 1254, 1089, 1071, 939, 831, 810,
2
–
1 1
7
3
3
3
77 cm . H NMR (300 MHz, room temp., CDCl ): δ = –0.20 (s,
tilled benzaldehyde (8, 305 µL, 3.0 mmol) at –78 °C. The reaction
mixture was stirred for 15 min, quenched with pyridine (5 drops)
H, Me), –0.01 (s, 3 H, Me), 0.71 (s, 6 H, Me), 0.77 (d, J = 6.9 Hz,
H, Me), 0.78 (d, J = 6.9 Hz, 3 H, Me), 1.51 (sept, J = 6.9 Hz, 1
followed by saturated aqueous NaHCO
temperature, and extracted with Et O (3×10 mL). The combined
organic layers were dried with Na SO , concentrated in vacuo, and
3
(5 mL), warmed to room
H, CH for thexyl group), 2.41 (s, 3 H, ArCH
5.0 Hz, 1 H, CH ), 3.55 (dd, J = 8.4, 15.0 Hz, 1 H, CH
dd, J = 3.9, 8.4 Hz, 1 H, CH), 7.23–7.42 (m, 7 H, Ar), 7.86 (d, J
8.1 Hz, 2 H, Ar-H for phenyl) ppm. C24 Si (382.64): calcd.
C 75.34, H 8.96; found C 75.32, H 8.99.
3
), 2.92 (dd, J = 3.9,
2
1
2
2
), 5.35
2
4
(
purified by flash column chromatography on silica gel (hexane/
EtOAc, 50:1) to give a mixture of 11, 15, 16, and 17 (1.15 g, Ͼ99%
yield). The ratio of products was determined as 1:1.0:0.89:0.81 by
=
34 2
H O
1
H NMR and GC analysis. The GC peaks were assigned on the
3
-(tert-Butyldimethylsilyloxy)-3-phenyl-1-(p-tolyl)propan-1-one (17):
To a solution of tert-butyldimethylsilyl trifluoromethanesulfonate
1.98 mL, 8.3 mmol) and triethylamine (2.1 mL, 15 mmol) in
CH Cl (20 mL) was added methyl 4-methylphenyl ketone (1.0 mL,
.5 mmol) at 0 °C. The reaction mixture was warmed to room tem-
perature, stirred for 3 h, quenched with a saturated aqueous
NaHCO solution (10 mL), and extracted with Et O (3×30 mL).
The combined organic layers were dried with Na SO , concentrated
+
+
basis of the [M – 15] or [M + H] peak as well as expected frag-
mentation patterns. GC/MS: tBuMe Si aldolate 11; t = 20.5 min,
Si aldolate 15; t = 22.8 min, m/z
Si aldolate 16; t = 22.5 min, m/z =
Si aldolate 17; t = 20.9 min, m/z = 355 [M
2
R
(
+
m/z = 341 [M + H] . ThexylMe
2
R
2
2
+
=
3
+
383 [M + H] . ThexylMe
2
R
7
+
53 [M – 15] . tBuMe
2
R
+
H] .
3
2
General Procedure for the Mukaiyama Aldol Reaction (Table 1): To
a solution of in situ prepared tBuMe SiNTf (0.03 mmol) in Et
2 mL) was added freshly distilled (1-cyclohexenyloxy)trimethylsil-
ane (18, 233 µL, 1.2 mmol) at –78 °C, followed by benzaldehyde (8,
2
4
2
2
2
O
in vacuo, and purified by flash column chromatography on silan-
ized silica gel (column wrapped with a dry ice jacket, pentane) to
give tert-butyldimethyl(1-p-tolylvinyloxy)silane (1.16 g, 63% yield)
as a colorless oil. IR (film): ν˜ = 2930, 2361, 1616, 1315, 1302, 1254,
(
1
2
02 µL, 1.0 mmol). The reaction mixture was stirred at –78 °C for
0 min, quenched with pyridine (5 drops) and saturated aqueous
–1
1
1
111, 1014, 836, 780 cm . H NMR (400 MHz, room temp.,
CDCl ): δ = 0.20 (s, 6 H, CH Si), 1.00 [s, 9 H, (CH )C], 2.34 (s, 3
H, ArCH ), 4.36 (d, J = 1.6 Hz, 1 H, CH ), 4.84 (d, J = 1.6 Hz, 1
H, CH ), 7.12 (d, J = 6.5 Hz, 2 H, m-H for p-tolyl group), 7.50 (d,
J = 6.5 Hz, 2 H, o-H for p-tolyl group) ppm. C NMR (100 MHz,
room temp., CDCl ): δ = –4.6 (CH Si), 21,2 [(CH CSi], 25.7
(CH CSi], 90.2 (CH ), 125.2 (o-CH for p-tolyl group), 128.7 (m-
CH for p-tolyl group), 135.0 [C–CSi(CH )], 138.0 (p-C for p-tolyl
NaHCO
bined organic layers were dried with Na
vacuo, and purified by flash column chromatography on silica gel
hexane/EtOAc, 50:1) to give a diastereomer mixture of the corre-
3
(5 mL), and extracted with Et
2
O (2×10 mL). The com-
3
3
3
2
SO , concentrated in
4
3
2
2
(
13
sponding silyl aldolate (277 mg, Ͼ99% yield, syn/anti = 62:38 de-
3
3
3 3
)
1
termined by H NMR). In all cases, the diastereoselectivities were
[
3
)
3
2
1
determined by H NMR of the diastereomer mixtures, according
2
to reported values.
group), 156.0 (p-tolylC–Si) ppm. The title compound was synthe-
sized according to the analoguous procedure for 15 using tert-bu-
tyldimethyl(1-p-tolylvinyloxy)silane instead of 14. TLC (hexane/
Dimethyl(thexyl)([1-(p-tolyl)vinyloxy]silane (14): To a solution of di-
methyl(thexyl)silyl chloride (3.5 mL, 18 mmol) and silver trifluoro-
methanesulfonate (4.2 g, 16.5 mL) in CH
methyl 4-methylphenyl ketone (2.0 mL, 15 mmol) and triethyl-
amine (4.2 mL, 30 mmol) at 0 °C. The reaction mixture was (300 MHz): δ = –0.18 (s, 3 H, Me), –0.09 (s, 3 H, Me), 0.75 (s, 9
warmed to room temperature, stirred for 3 h, quenched with a satu- H, tBu), 2.41 (s, 3 H, ArCH ), 2.92 (dd, J = 3.9, 15.3 Hz, 1 H,
rated aqueous NaHCO solution (10 mL), and extracted with Et CH ), 3.55 (dd, J = 8.7, 15.3 Hz, 1 H, CH ), 5.35 (dd, J = 3.9,
3×50 mL). The combined organic layers were dried with Na SO 8.7 Hz, 1 H, CH), 7.23–7.43 (m, 7 H, Ar), 7.87 (d, J = 8.1 Hz, 2
2
Cl
2
(20 mL) was added
EtOAc, 4:1): R
f R
= 0.74. GC: t = 20.9 min. IR (film): ν˜ = 2955,
–
1 1
1686, 1607, 1256, 1092, 1071, 939, 835, 808, 779 cm . H NMR
3
3
2
O
2
2
(
2
4
,
concentrated in vacuo, and purified by flash column chromatog-
raphy on silanized silica gel (column wrapped with a dry ice jacket,
hexane/EtOAc, 50:1) to give 14 (3.36 g, 81% yield) as a colorless
oil. IR (film): ν˜ = 2959, 1614, 1511, 1465, 1314, 1301, 1183, 1111,
30 2
H, Ar-H for phenyl) ppm. C22H O Si (354.56): calcd. C 74.53, H
8.53; found C 74.67, H 8.52.
–1 1
832, 781 cm . H NMR (300 MHz, room temp., CDCl
3
): δ = 0.23 Acknowledgments
(
s, 6 H, Me), 0.95 (m, 12 H, Me), 1.74 (m, 1 H, CH for thexyl
group), 2.35 (s, 3 H, ArCH ), 4.35 (d, J = 1.5 Hz, 1 H, CH ), 4.83
d, J = 1.5 Hz, 1 H, CH
), 7.13 (d, J = 8.7 Hz, 2 H, m-H for p- the Japan Science and Tecnology Corp. (JST), National Institute
tolyl group), 7.50 (d, J = 8.7 Hz, 2 H, o-H for p-tolyl group) ppm.
of Health (NIH) GM068433-01, and a starter grant from the Uni-
Si), 18.5 versity of Chicago.
Ph), 25.1 [SiC(CH CH],
), 125.2 (o-CH for p-tolyl group), 128.7
m-CH for p-tolyl group), 135.2 [C–CSi(CH )], 137.9 (p-CH for p-
tolyl group), 155.9 [p-tolylC–Si(CH )] ppm.
Support for this research was provided by the SORST project of
3
2
(
2
1
3
C NMR (100 MHz, room temp., CDCl
(CH CSi], 20.2 [(CH CH], 21.2 (CH
3.9 [(CH CH], 90.3 (CH
3 3
): δ = –2.7 (CH
[
3
3
)
2
3
)
2
3
3 2
)
3
)
2
2
[
1] E. M. Carreira, in: Comprehensive Asymmetric Catalysis III
Ed.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto), Springer-Ver-
lag, Berlin, Heidelberg, 1999, p. 997.
(
2
(
2
3
-[Dimethyl(thexyl)silyloxy]-3-phenyl-1-(p-tolyl)propan-1-one (15): [2] E. M. Carreira, R. A. Singer, Tetrahedron Lett. 1994, 35, 4323–
To a solution of in situ prepared Me
3
SiNTf
2
(ca. 0.005 mmol) and
4326.
Eur. J. Org. Chem. 2006, 1837–1844
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1843

FULL PAPER
[
[
[
4
mediate and 5a or 5b:W. W. Ellis, B. Bosnich, Chem. Commun.
1998, 193–194.
[15] K. Ishihara, Y. Hiraiwa, H. Yamamoto, Chem. Commun. 2002,
1564–1565.
[
6] B. Mathieu, L. Ghosez, Tetrahedron Lett. 1997, 38, 5497–5500.
7] A. Ishii, O. Kotera, T. Saeki, K. Mikami, Synlett 1997, 1145–
[16] These scrambling experiments exclude the possibility for path
C and D in Scheme 1.
[17] D. D. Dilman, S. L. Loffe, Chem. Rev. 2003, 103, 733–772.
[18] K. Ishihara, Y. Hiraiwa, H. Yamamoto, Synlett 2000, 80–82.
[19] S. Murata, M. Suzuki, R. Noyori, Tetrahedron 1988, 44, 4259–
4275.
[20] M. B. Berry, R. J. Griffiths, J. A. Howard, J. M. A. Leech, P. G.
Steel, D. S. Yufit, J. Chem. Soc., Perkin Trans. 1 1991, 3645–
3650.
[
1
146.
8] N. Kuhnert, J. Peverley, J. Robertson, Tetrahedron Lett. 1998,
9, 3215–3216.
[
3
[
[
9] J. Cossy, F. Lutz, V. Alauze, C. Meyer, Synlett 2002, 45–48.
10] K. Ishihara, Y. Hiraiwa, H. Yamamoto, Synlett 2001, 1851–
1854.
[11] M. G. Davidson, P. R. Raithby, A. L. Johnson, P. D. Bolton,
Eur. J. Inorg. Chem. 2003, 3445–3452.
[21] A. Vij, Y. Y. Zheng, R. L. Kirchmeier, J. M. Shreeve, Inorg.
Chem. 1994, 33, 3281–3288.
[
[
12] L. Turowsky, K. Seppelt, Inorg. Chem. 1988, 27, 2135–2137.
13] F. J. Waller, A. G. M. Barrett, D. C. Braddock, D. Ramprasad,
R. M. McKinnell, A. J. P. White, D. J. Williams, R. Ducray, J.
Org. Chem. 1999, 64, 2910–2913.
Received: October 29, 2005
Published Online: February 9, 2006
1844
www.eurjoc.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2006, 1837–1844