Platinum-catalyzed enantioselective aldol addition of ketene silyl acetals to aldehydes

Article abstract of DOI:10.1021/ja981970d

The first platinum-catalyzed enantioselective aldol reactions are reported. The catalysts are generated from chiral bisphosphine platinum acyl complexes by activation with noncoordinating or weakly coordinating strong acids and can be used to catalyze the

Full text of DOI:10.1021/ja981970d

10032
J. Am. Chem. Soc. 1998, 120, 10032-10039
Platinum-Catalyzed Enantioselective Aldol Addition of Ketene Silyl
Acetals to Aldehydes
Osamu Fujimura
Contribution from the AdVanced Organic Chemical Research Group, Ube Laboratory,
Corporate Research and DeVelopment, Ube Industries, Ltd., Ube City, Yamaguchi 755-8633, Japan
ReceiVed June 5, 1998
Abstract: The first platinum-catalyzed enantioselective aldol reactions are reported. The catalysts are generated
from chiral bisphosphine platinum acyl complexes by activation with noncoordinating or weakly coordinating
strong acids and can be used to catalyze the enantioselective aldol addition of ketene silyl acetals to various
aldehydes. The presence of oxygen and water during catalyst activation is required to obtain enantioselectivity.
The enantioselectivities observed in the cases of aliphatic primary aldehydes are more than 90% enantiomer
excess. IR and 31P NMR spectroscopies were used to demonstrate the intermediacy of C-bound platinum
enolate in the aldol process.
Introduction
precursors.4,8 For example, there are excellent works utilizing
copper(II) complexes as Lewis acids reported by Evans9 and a
recent report by Carreira regarding the copper(II) enolate
mediated asymmetric addition of dienolate to aldehydes.5 As
part of an ongoing project directed toward the development of
useful catalysts for potential industrial applicability, we inves-
tigated the possibilities of platinum(II) cationic complexes
serving as either Lewis acids or metal enolate precursors for
catalytic enantioselective aldol reactions. Although cationic Pt
complexes are known to behave as very strong Lewis acids,10
there are no examples of the application of these complexes to
reactions mediated by Lewis acids. In this paper, we report
the first example of platinum-catalyzed enantioselective aldol
reactions, which are considered to be mediated by platinum
C-enolates. The scope and limitations of this new process will
be discussed.
Enantioselective aldol addition reactions are one of the most
important methodologies utilized for the construction of asym-
metric carbon-carbon bonds.1 Two major approaches have
been developed for the catalytic variants of this reaction. One
is to employ chiral Lewis acids to activate the carbonyl
compounds toward the addition of preformed latent enolates
such as enol silanes. The other is to utilize chiral transition
metal complexes and latent enolates to generate transition metal
enolates as catalytic intermediates. Concerning the former
approach, significant progress has been made in recent years.2,3
On the other hand, the number of examples reported on aldol
addition reactions involving transition metal enolates is rather
limited.4,5
Late transition metal complexes generally show remarkable
activities as Lewis acids (once they are converted to cationic
form)6,7 and may also have the potential to be metal enolate
Results and Discussion
(1) For a review of asymmetric aldol addition reactions, see: (a) Group
I and II enolates: Heathcock, C H. In ComprehensiVe Organic Synthesis;
Trost, B. M., Fleming, I., Heathcock C. H., Eds.; Pergamon: New York,
1991; Vol. 2, pp 181-238. (b) Group III enolates: Kim, B. M.; Williams,
S. F.; Masamune, S. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Heathcock, C. H., Eds.; Pergamon: New York, 1991; Vol. 2,
pp 239-275. (c) Transition metal enolates: Paterson, I. In ComprehensiVe
Organic Synthesis; Trost, B. M., Fleming, I., Heathcock C. H., Eds.;
Pergamon: New York, 1991; Vol. 2, pp 301-319.
Preparation of the Catalyst Precursors. Platinum cationic
species such as platinum triflate(II) or platinum tetrafluoborate-
(II) are generally derived from K2PtCl4 in three steps:11 (1)
synthesis of Pt(diene)Cl2 complexes; (2) replacement of the
dienes by the desired ligands (like phosphines); and (3) halogen
exchange reaction with silver salts of noncoordinating or weakly
coordinating strong acids (Scheme 1). This multistep synthesis
is relatively easy to perform on a laboratory scale, but several
difficulties arise during large scale production. In particular,
the stoichiometric use of silver salts in the last step raise the
cost of catalysts for industrial processes. (The product AgCl
can be recovered and converted to AgOTf or AgBF4, but these
additional processes also result in higher costs.)
(2) For a review of Lewis acid-catalyzed asymmetric aldol additions of
latent enolates, including Lewis basic catalysts, see: Nelson, S. G.
Tetrahedron: Asymmetry 1998, 9, 357-389.
(3) Recently, direct catalytic asymmetric aldol reactions were reported,
see: Yamada, Y. M. A.; Yoshikawa, N.; Sasai, H.; Shibasaki, M. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1871-1873.
(4) Palladium: (a) Hagiwara, E.; Fujii, A.; Sodeoka, M. J. Am. Chem.
Soc. 1998, 120, 2474-2475. (b) Sodeoka, M.; Tokunoh, R.; Miyazaki, F.;
Hagiwara, E.; Shibasaki, M. Synlett 1997, 463-466. (c) Sodeoka, M.; Ohrai,
K.; Shibasaki, M. J. Org. Chem. 1995, 60, 2648-2649.
(5) Copper: Kru¨ger, J.; Carreira, E. M. J. Am. Chem. Soc. 1998, 120,
837-838.
(6) (a) Corey, E. J.; Imai, N.; Zhang, H. Y. J. Am. Chem. Soc. 1991,
113, 728-729. (b) Khiar, N.; Fernandez, I.; Alcudta, T. Tetrahedron Lett.
1993, 34, 123-126.
(7) (a) Evans, D. A.; Murry, J. A.; Matt, P.; Norcross, R. D.; Miller, S.
J. Angew. Chem., Int. Ed. Engl. 1995, 34, 798-800. (b) Evans, D. A.; Miller,
S. J.; Lectka, T. J. Am. Chem. Soc. 1993, 115, 6460-6461. (c) Evans, D.
A.; Lectka, T.; Miller, S. J. Tetrahedron Lett. 1993, 34, 7027-7030.
(8) Slough, G. A.; Bergman, R. G.; Heathcock, C. H. J. Am. Chem. Soc.
1989, 111, 938-949.
(9) (a) Evans, D. A.; Kozlowski, M. C.; Burgey, C. S.; MacMillan, D.
W. C. J. Am. Chem. Soc. 1997, 119, 7893-7894. (b) Evans, D. A.; Murry,
J. A.; Kozlowski, M. C. J. Am. Chem. Soc. 1996, 118, 5814-5815.
(10) (a) Longato, B.; Corain, B.; Bonora, G. M.; Pilloni, G. Inorg. Chim.
Acta 1987, 137, 75-79. (b) Longato, B.; Pillioni, G.; Valle, G.; Corain, B.
Inorg. Chem. 1988, 27, 956-958.
(11) Stang, P. J.; Cao, D. H.; Saito, S.; Arif, A. M. J. Am. Chem. Soc.
1995, 117, 6273-6283. See also ref 10.
S0002-7863(98)01970-2 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/19/1998

Pt-Catalyzed EnantioselectiVe Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 120, No. 39, 1998 10033
Scheme 1
lutidine was added as a proton scavenger.14,15 The complexes
did not catalyze the reaction without activation by triflic acid.
Although the reaction proceeded cleanly and very effic-
iently, the enantiomer excess observed for the product was 0%.
After examining various reaction and activation conditions,
we discovered that the conditions under which 1 was activated
have significant effects on asymmetric induction. For example,
when the activation of complex 1a was conducted in the
presence of oxygen (i.e. air), the reaction proceeded to yield a
product with an enantiomer excess of 35%! Furthermore,
addition of 2 equiv of water to the Pt complex in the presence
of oxygen increased the enantiomer excess to 59%.
When hydrocinnamaldehyde was used as substrate, only 16%
ee was observed for the reaction product when the catalyst was
activated under Ar in dry CH2Cl2. This was improved to 77%
ee when the activation was carried out under air in dry CH2-
Cl2, and an enantiomer excess of 95% was realized when the
activation was carried out under air in the presence of 2 equiv
of water. Water itself improved the enantioselectivity to a small
extent (3-5% ee), but excess water resulted in the deactivation
of the catalyst.
To avoid these problems, we selected platinum(II) acyl
complexes derived from salicylaldehyde and 2 equiv of tertiary
phosphines as candidates for the catalyst precursors. These
chelating acyl platinum(II) complexes were first reported by
Pregosin12 and possess several attractive features. First, they
can be synthesized from commercially available starting materi-
als in a single step. Second, they can be easily handled due to
their stability toward air and moisture. Finally, the correspond-
ing platinum cationic species are easily generated by the
treatment with noncoordinating or weakly coordinating strong
acids.13 Despite these attractive features, applications of these
complexes to organic synthesis have been primarily limited to
the oxidation reactions developed by Strukul and co-workers.13
We anticipated that these complexes, once transformed to a
cationic species, would be active catalysts for enantioselective
aldol reactions as either Lewis acids or metal enolate precursors.
To investigate these possibilities, a variety of Pt complexes were
synthesized (Figure 1, Scheme 2).
The presence of oxygen is only required in the catalyst
activation stage, and replacement of the reaction atmosphere
by Ar after activation slightly improved the enantiomer excess
of the product (3-5% ee). Activation of the catalyst under Ar
followed by exposure to air resulted in the loss of reactivity as
well as the loss of enantioselectivity.
After establishment of the optimized activation conditions, a
series of complexes are screened with benzaldehyde as a
substrate, which enabled rapid screening (Table 1, entries 1 and
3-6). The best results were obtained with 1a and 1b. The
difference in the substitution patterns on the salicyl group
seemed to have almost no effect on the reaction. Various
aldehydes were also screened as reaction substrates with 1a and
1b as catalyst precursors. The results are shown in Table 1.
Aromatic and R,â-unsaturated aldehydes showed almost the
same reactivities and degrees of enantioselectivities. Primary
aliphatic aldehydes showed high enantioselectivities around 90%
ee. Though the reaction rate became slower, a 70-80%
conversion of the starting aldehydes was observed in 20 h at
-25 °C, and around 90% conversion was observed in 40 h. On
the contrary, secondary and tertiary aldehydes showed almost
no reactivity under these conditions.
The reaction temperature has significant effects on the
reaction rate and enantioselectivity. Reactions were faster at 0
°C, but enantioselectivity diminished more than 20% in enan-
tiomer excess (entries 2 and 10). Reactions were too sluggish
to proceed at -78 °C.
As previously mentioned, addition of base to scavenge trace
amounts of acid is critical in maintaining the high enantiomer
excess. Without addition of base, the reaction proceeded very
rapidly, but resulted in diminished enantioselectivity (entry 11).
It is most likely that a catalytic amount of trimethylsilyltriflate
was formed and catalyzed the nonstereoselective aldol reac-
tion.16,17
Figure 1. Catalyst precursors (isolated yields are shown in parenthe-
ses).
Enantioselective Aldol Addition of Ketene Silyl Acetals
to Aldehydes. When complex 1a (5 mol %) was activated by
the addition of HOTf (5 mol %) in dry CH2Cl2 under Ar, the
resulting solution catalyzed the reaction between methyl tri-
methylsilyl dimethylketene acetal and benzaldehyde at -25 °C
in 16 h. The desired aldol product was obtained in 99% yield.
We chose triflic acid because it can be easily handled and
obtained as neat. To avoid the influences of residual acid, 2,6-
(14) (a) Singer, R. A.; Carreira, E. M. J. Am. Chem. Soc. 1995, 117,
12360-12361. (b) Carreira, E. M.; Lee, W.; Singer, R. A. J. Am. Chem.
Soc. 1995, 117, 3649-3650.
(15) HOTf (5.0 mol %) catalyzed the reaction to be completed at -78
°C in 30 min.
(16) The aldol addition of ketene silyl acetals to aldehydes is catalyzed
by silyl triflates. See: Carreira, E. M.; Singer, R. A. Tetrahedron Lett. 1994,
35, 4323-4326.
(17) Similer formation of trimethylsilyl tetrafluoborate is reported in ref
4a.
(12) (a) Anklin, C.; Pregosin, P. S.; Bachechi, F.; Mura, P.; Zambonelli,
L. J. Organomet. Chem. 1981, 222, 175-185. (b) Motschi, H.; Pregosin,
P. S.; Ruegger, H. J. Organomet. Chem. 1980, 193, 397-405.
(13) (a) Baccin, C.; Gusso, A.; Pinna, F.; Strukul, G. Organometallics
1995, 14, 1161-1167. (b) Gusso, A.; Baccin, C.; Pinna, F.; Strukul, G.
Organometallics, 1994, 13, 3442-3451.

10034 J. Am. Chem. Soc., Vol. 120, No. 39, 1998
Fujimura
Table 1. Platinum-Catalyzed Enantioselective Aldol Addition of Ketene Silyl Acetal to Aldehydesa
entry
RCHO
PhCHO
catalyst
temp (°C)/time (h)
yield (%)b
R′:SiMe3/Hc
eed-f
1
2
3
4
5
6
7
8
9
10
11g
12
13
14
15
16
17
18
19
20
1a
-25/21
99 (99)
99 (99)
99 (99)
97 (95)
99 (99)
82 (85)
90 (86)
92 (94)
94 (96)
95 (96)
94 (93)
96 (95)
92 (87)
99 (98)
96 (94)
66 (77)
(7)
65/35
82/18
82/18
75/25
71/29
86/14
65/35
84/16
51/49
90/10
72/28
84/16
55/45
91/9
59
30
56
26
41
10
46
46
95
74
43
91
91
90
88
80
0/19
1b
1c
1d
1e
1a
1b
1a
-25/19
-25/17.5
-25/20.5
-25/27
(E)-PhCHdCHCHO
-25/144
-25/143
-25/168
0/17
-78 to -25/5
-25/170
-25/140
-25/144
-25/171
-25/171
-25/98
PhCH2CH2CHO
1b
1a
1b
1a
1b
1a
1b
1a
1b
(CH3)2CHCH2CHO
(CH3)3CCH2CHO
cC6H11CHO
82/18
94/6
65/35
>99/1
NDh
NDh
NDh
NDh
-25/98
-25/120
-25/120
(5)
t-BuCHO
NRi
NRi
a Catalyst activation and substrates addition was performed under air, then the reaction was performed under Ar (H2O equivalent to Pt ) 2.0).
b Isolated yield, yields in parentheses are determined by GC. c Determined by GC. d Observed ee values for R′ ) SiMe3 and R′ ) H were identical.
e ee values for R′ ) SiMe3 were determined by chiral HPLC (CHIRALPAK-AD, Daicel Co.) after deprotection by TBAF. f ee values for R′ ) H
were determined by chiral HPLC (CHIRALPAK-AD, Daicel Co.). g Lutidine was not added. h Not determined. i No reaction.
Scheme 2. Preparation of Catalyst Precursors
three NMR experiments were conducted with complex 1a.
Sample a was prepared in dry CD2Cl2 and the activation process
was carried out under an N2 atmosphere. Sample b was
prepared in dry CD2Cl2 and the activation process was carried
out in an atmosphere of air. Sample c was prepared in wet
CD2Cl2 (containing 2 equiv of water to complex 1a) and the
activation process was carried out under air.
Very rapid conversion of starting complex 1a was observed
in sample c and evolution of a new phosphine-platinum species
(complex A) was observed by 31P NMR at δ 3.25 (JPt-P ) 3674
Hz). Figure 2 shows 31P NMR spectra of 1a and complex A.
In sample b, the conversion of starting 1a is much slower but
after a long period of time (22 h) it became almost identical to
the spectrum of sample c. In the case of sample a, the
conversion of the starting 1a was very slow and complex A,
observed in the case of samples b and c, was never observed.
NMR samples b and c were used in sample reactions. Reactions
proceeded without any problem with these NMR sample
solutions and nearly the same reactivity and enantioselectivity
were observed. We have concluded that platinum phosphine
complex A is the actual catalytic species, that the presence of
oxygen is necessary for its formation, and that the presence of
water accelerates the rate of its formation.18 The salicylaldehyde
part of 1a was transformed and observed as 2,4-di-tert-butyl-
phenol by 1H NMR, consistent with the activation scheme shown
in Scheme 4 (as suggested by Strukul and co-workers).13
The exact nature of complex A is still unknown, since we
could not isolate the complex. However, judging from the Pt-P
coupling constant and symmetric nature of the complex (only
one phosphorus peak was observed), we assume that complex
A has an OH group in the position trans to the phosphorus.19
We synthesized two known platinum cationic complexes [Pt-
((R)-BINAP)(H2O)][OTf]2 (2)20 and [Pt((R)-BINAP)(µ-OH)]2-
(BF4)2(3),19,21 which are considered to be similar to complex
Scheme 3. Chemoselective Asymmetric Aldol Reaction
Taking advantage of the reactivity differences outlined above,
a chemoselective asymmetric aldol addition was performed. In
the presence of both primary and secondary aliphatic aldehydes,
only the primary aldehyde was selectively reacted to give the
desired aldol product in 94% ee and 96% yield, while the
secondary aldehyde was recovered unchanged (Scheme 3).
NMR Studies and Mechanistic Considerations. NMR
studies were conducted, together with the synthesis and reactiv-
ity trials of known authentic platinum cationic complexes, to
elucidate the nature of the reactive species involved in this
system.
(18) The exact role of oxygen and water is still under investigation.
(19) Strukul, G.; Varagnolo, A.; Pinna, F. J. Mol. Catal. A: Chem. 1997,
117, 413-423.
(1) Effect of O2 and Water on the Activation of Complex
1. To observe the activation process of complex 1 by TfOH,

Pt-Catalyzed EnantioselectiVe Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 120, No. 39, 1998 10035
1
observed by H and 13C NMR, and no change was observed
for complex A by 31P NMR. If complex A has a Lewis acidic
nature, it is likely that a change would be observed for one or
more of these chemical shifts. Therefore, it is unlikely that
complex A is acting as a Lewis acid.
Next, methyl trimethylsilyl dimethylketene acetal (KSA) was
added to the CD2Cl2 solution of complex A at room temperature.
With addition of up to 3 equiv of KSA, no change in complex
A was observed by 1H or 31P NMR and little hydrolysis of the
KSA was observed. But the addition of excess KSA (10 equiv)
resulted in the formation of a new Pt-phosphine complex
(complex B) in the 31P NMR spectrum that has two nonequiva-
lent phosphorus shifts at δ 26.8 and 9.5. These resonances have
a common phosphorus-phosphorus coupling constant (JP-P
)
19 Hz) and platinum-phosphorus coupling constants of JPt-P
) 1904, and 4789 Hz, respectively. The amount of complex
B increased over time and after addition of another 18 equiv of
KSA, complex A was completely consumed and exclusive
formation of complex B was observed (Figure 3). 1H NMR
was not informative as many small ambiguous peaks started to
appear.
It is reasonable to assume that complex B is a platinum
enolate, as this is formed by the reaction of complex A and
KSA (excess). However, there could be possibilities of the
O-bound enolate (4a,b), or the C-bound enolate (5a,b) for
complex B (Figure 4).
In the case of the O-bound enolate, the JPt-P observed for
the trans position of the enol alkoxide should be larger than
4000 Hz.24 As the other coordination site is considered to be
occupied by OH or the chelating methoxy group of the enolate,
the other JPt-P also should be larger than 3700 Hz.25 In this
way, the assumed structure for the O-bound enolate could not
account for the 31P NMR spectrum of complex B. On the other
hand, in the case of the C-bound enolate, the JPt-P observed
for the δ 26.8 peak (1904 Hz) is within the typical JPt-P
values for the phosphorus trans to the Pt-C σ-bond (1800-
2000 Hz).26,27 The other peak at δ 9.5 (JPt-P ) 4789 Hz) could
be the other phosphorus trans to OH or chelating carbonyl
group. According to the 31P NMR data for platinum homo-
enolate, C-bound enolate,26 and BINAP dimethyl complex,27
we feel it reasonable to assume that complex B is C-bound
enolate 5a or 5b.
In addition, we observed a strong infrared stretching absorp-
tion at 1594 cm-1, which is consistent with the organic carbonyl
absorption strongly chelating to platinum, supporting the C-
bound enolate structure 5b.26,28
The proposed catalytic cycle of the asymmetric reaction is
shown in Scheme 5. The requirement for excess KSA to
generate complex B for the successful enantioselective aldol
reaction is also confirmed by the following experiments.
When the reaction was carried out with 2 equiv of hydro-
cinnamaldehyde and 2.8 equiv of methyl trimethylsilyl di-
methylketene acetal, the reaction did not proceed even with
Figure 2. 31P NMR of 1a in CD2Cl2 (top) and complex A (bottom).
Scheme 4
A, and tested their reactivities in aldol addition reactions. The
results are shown in Table 2.
Complex 2 catalyzed the reaction efficiently, but the resultant
product was racemic. Complex 3 catalyzed the reaction but
with a much lower activity and a low level of enantioselectivity.
Thus, complex A appears to be a different platinum species.
There is a possibility that the phosphine ligand (BINAP) was
oxidized to form the corresponding phosphine oxides (i.e.
BINAP dioxide22 or BINAP monoxide23); however, they were
not observed by the 31P NMR monitoring of the activation
process.
(2) The Role of Platinum Complex A as a Lewis Acid or
a Platinum Enolate Precursor. To elucidate the role of
complex A, 3 equiv of hydrocinnamaldehyde were added to
the CD2Cl2 solution of complex A at room temperature.
However, no change of the chemical shifts of the aldehyde was
(24) The phosphorus trans to alkenoxy (aryloxy) groups have JPt-P of
4100-4600 Hz. See refs 12b and 13b.
(25) The typical JPt-P value for the phosphorus trans to hydroxyl group
is 3600-4000 Hz. See ref 19 and related references cited therein.
(26) Mhinzi, G. S.; Crascall, L. E.; Spencer, J. L. Inorg. Chim. Acta
1997, 265, 83-87.
(27) Nozaki, K.; Sato, N.; Tonomura, Y.; Yasutomi, M.; Takaya, H.;
Hiyama, T.; Matsubara, T.; Koga, N. J. Am. Chem. Soc. 1997, 119, 12779-
12795.
(28) After confirmation of the formation of complex B, solvents, excess
KSA, and hydrolyzed methyl isobutyrate were removed in vacuo and
residual solid was analyzed by IR. Very weak absorption at 1605 cm-1
from 2,4-di-tert-butylphenol seemed to be overlapped by this strong CdO
absorption.
(20) Stang, P. J.; Olenyuk, B.; Arif, A. M. Organometallics 1995, 14,
5281-5289.
(21) Since a very small amount of µ-OH by integration compared with
1
aromatic protons was observed by H NMR of complex A at δ ) -0.65
ppm, the structure of A is not the µ-OH bridged binuclear platinum hydroxy
complex like 3.
(22) Takaya, H.; Mashima, K.; Koyano, K.; Yagi, M.; Kumobayashi,
H.; Taketomi, T.; Akutagawa, S. J. Org. Chem. 1986, 51, 629-635.
(23) Ozawa, F.; Kubo, A.; Hayashi, T. Chem. Lett. 1992, 2177-2180.

10036 J. Am. Chem. Soc., Vol. 120, No. 39, 1998
Table 2. Aldol Reaction by Platinum Cation Complexes
Fujimura
entry
catalyst
additive
temp (°C)/time (h)
yield (%)a
R′:SiMe3/Hb
eec-e
1
2
3
4
[Pt((R-BINAP)(H2O)][OTf]2
none
-78/0.5
-25/17
-25/18
-25/20
99 (99)
99 (99)
27 (32)
11 (10)
91/9
98/2
54/44
52/48
4
4
16
20
lutidine (5.0 mol %)
none
lutidine (5.0 mol %)
[Pt((R-BINAP)(µ-OH)]2(BF4)2
a Isolated yield, yields in parentheses are determined by GC. b Determined by GC. c Observed ee values for R′ ) SiMe3 and R′ ) H were
identical. d ee values for R′ ) SiMe3 were determined by chiral HPLC (CHIRALPAK-AD, Daicel Co.) after deprotection by TBAF. e ee values for
R′ ) H were determined by chiral HPLC (CHIRALPAK-AD, Daicel Co.).
H2O concentration in organic solvents was determined by a Kyoto
Electronics MKC-210 Karl Fischer moisture titrator.
All manipulations involving air-sensitive materials were performed
under an argon atmosphere with standard Schlenk techniques with a
double-manifold vacuum line or in a N2 filled drybox, using oven-
dried glassware with magnetic stirring.
catalyst loadings as high as 50 mol %. GC analysis showed
no conversion of either aldehyde or ketene silyl acetal. By
increasing the KSA amount to 28 equiv, reaction ensued and
after 69 h yielded the product in quantitative yield and 95% ee.
Conclusion
Materials. Commercial reagents were used as received, with the
following exceptions. Diethyl ether and THF were vacuum transferred
from sodium-benzophenone ketyl, degassed by freeze-pump-thaw
(FPT) three times, and stored under argon in a Schlenk flask.
Dichloromethane was dried over CaH2, vacuum transferred to a Schlenk
flask, degassed by FPT, and stored under argon. 2,6-Lutidine was dried
over CaH2, distillated to a Schlenk flask, degassed by FPT, and stored
under argon. DMSO was dried over MS4A, filtered, distillated to a
Chiral bisphosphine platinum acyl complexes generate useful
catalysts for the enantioselective aldol addition of ketene silyl
acetal to various aldehydes via activation with noncoordinating
or weakly coordinating strong acids. The presence of oxygen
and water significantly improves the enantioselectivity and in
the cases of aliphatic primary aldehydes, an enantiomer excess
of more than 90% is realized. The reaction is sensitive to the
substitution pattern of the substrate aliphatic aldehydes and only
primary aldehydes are capable of reaction with the current
system. The reaction is proposed to be mediated by a C-bound
platinum enolate, based on 31P NMR and IR spectroscopic data.
The platinum species are not acting as Lewis acids. Further
studies concerning the reaction intermediates, the addition mode
of platinum-enolate and aldehydes, the improvement of enan-
tiomer excess, and the development of improved catalyst
systems are currently underway.
20
Schlenk flask, and stored under argon. [Pt((R)-BINAP)(H2O)][OTf]2
19
and [Pt((R)-BINAP)(µ-OH)]2(BF4)2 were prepared according to the
literature.
Pt{O(3,5-tBu2C6H2)CO}{(R)-BINAP}; [Pt{(R)-BINAP}(3,5-DT-
BS)] (1a). To a 50 mL Schlenk flask were added K2PtCl4 (0.67 g, 1.6
mmol), Na2CO3 (0.81 g, 4.8 mmol), 3,5-di-tert-butyl-2-hydroxyben-
zaldehyde (0.38 g, 1.6 mmol), and DMSO (25 mL) under argon. The
reaction mixture was heated to 140 °C. After the mixture was stirred
for 40 min at the same temperature, the resulting yellow green
suspension was slowly cooled to 100 °C and (R)-BINAP (1.0 g, 1.6
mmol) was added. The yellow suspension was further cooled to 60
°C and DMSO was removed in vacuo. The dry residue was extracted
by CH2Cl2 and filtered. The yellow filtrate was reduced in volume to
a few milliliters, followed by the addition of EtOH. The yellow
microcrystals were precipitated out and collected, washed by EtOH,
and dried in vacuo at 50 °C for 3 h (yield of 1, 1.4 g, 83%).
Pt{O(3,5-tBu2C6H2)CO}{(R)-BINAP} (1a): IR (KBr) 1623, 1437,
Experimental Section
General Methods. 1H NMR spectra were recorded on a JEOL
GX270 spectrometer or a EX400 spectrometer at ambient temperature.
Data are reported as follows: chemical shift in parts per million
downfield from tetramethylsilane (δ scale) with the solvent resonance
employed as an internal standard (CD2Cl2 at δ 5.31, C6D6 at δ 7.15,
CDCl3 at δ 7.26), multiplicity (br ) broad, s ) singlet, d ) doublet,
t ) triplet, q ) quartet, and m ) multiplet), coupling constant (Hz),
integration, and assignment.
1
1426, 744, 696, 523 cm-1; H NMR (400 MHz, CDCl3) δ 7.91-7.86
(m, 4H, aromatic), 7.55-6.98 (m, 22H, aromatic), 6.75-6.72 (m, 4H,
aromatic), 6.60-6.57 (m, 4H, aromatic), 1.20 (s, 9H, tBu), 1.03 (s,
9H, tBu); 31P NMR (160 MHz, CDCl3) δ 23.7 (d, JP-P ) 10.7, JPt-P
)
13C NMR spectra were recorded on a JEOL GX270 spectrometer or
a EX400 spectrometer at ambient temperature. 13C NMR chemical
shifts are reported in parts per million downfield from tetramethylsilane
(δ scale) with the solvent resonance employed as an internal standard
(C6D6 at δ 128.0, CDCl3 at δ 77.0). All 13C NMR spectra were
determined with complete proton decoupling.
1512, trans to C), 20.4 (d, JP-P ) 10.7, JPt-P ) 4474, trans to O);
[R]25 +582° (c 0.44, CH2Cl2). Anal. Calcd for C59H52O2P2Pt: C,
D
67.48; H, 4.99. Found: C, 66.38; H, 4.94.
Pt(OC6H4CO){(R)-BINAP}; [Pt{(R)-BINAP}(Sal)] (1b). The
procedure used for the preparation of complex 1a was followed except
salicylaldehyde (0.20 g, 1.6 mmol) was used instead of 3,5-di-tert-
butyl-2-hydroxybenzaldehyde (0.38 g, 1.6 mmol), giving 1.3 g (88%)
of complex 1b as yellow orange microcrystals.
31P NMR spectra were recorded on a JEOL EX400 spectrometer at
ambient temperature. All 31P NMR were determined with complete
proton decoupling and are reported in parts per million relative to
external 85% H3PO4 (δ scale).
Pt(OC6H4CO){(R)-BINAP} (1b): IR (KBr) 1627, 1456, 1436, 877,
1
743, 695 cm-1; H NMR (400 MHz, CDCl3) δ 8.07-7.97 (m, 4H,
Infrared spectra were obtained on a BIO-RAD Laboratories FTS-
65A FTIR. High-resolution mass spectra were provided by UBE Scien-
tific Analysis Laboratory. Analytical thin-layer chromatography was
accomplished with EM Reagents 0.25 mm silica gel 60 plates. Flash
chromatography was performed on Wakogel (silica gel) 200. Elemental
analysis was performed at UBE Scientific Analysis Laboratory. Optical
rotations were obtained on a JASCO DIP-370 digital polarimeter.
GLC analysis was performed on a Shimadzu GC-14A gas chro-
matograph.
aromatic), 7.53-7.25 (m, 17H, aromatic), 7.13-6.98 (m, 5H, aromatic),
6.79-6.57 (m, 9H, aromatic), 6.36 (dd, J ) 7.3, 7.3, 1H, aromatic);
31P NMR (160 MHz, CDCl3) δ 24.4 (d, JP-P ) 11.6, JPt-P ) 1491,
trans to C), 19.2 (d, JP-P ) 11.6, JPt-P ) 4536, trans to O). Anal.
Calcd for C51H36O2P2Pt: C, 65.31; H, 3.87. Found: C, 64.88; H, 3.96.
Pt{O(3,5-tBu2C6H2)CO}{(R)-Tol-BINAP}; [Pt{(R)-Tol-BINAP}-
(3,5-DTBS)] (1c). The procedure used for the preparation of complex
1a was followed, starting from K2PtCl4 (0.31 g, 0.74 mmol), Na2CO3
(0.23 g, 2.2 mmol), and 3,5-di-tert-butyl-2-hydroxybenzaldehyde (0.17
g, 0.74 mmol), and (R)-Tol-BINAP (0.50 g, 0.74 mmol) was used
instead of (R)-BINAP, giving 0.59 g (73%) of complex 1c as yellow
microcrystals.
HPLC analysis was performed on a Shimadzu LC10 high perfor-
mance liquid chromatograph with a chiral column (CHIRALPAK AD,
0.46 cm φ X 25 cm, Daicel Chemical Co.).

Pt-Catalyzed EnantioselectiVe Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 120, No. 39, 1998 10037
1.20 (s, 9H, tBu), 1.01 (s, 9H, tBu); 31P NMR (160 MHz, CDCl3) δ
21.8 (d, JP-P ) 10.4, JPt-P ) 1517, trans to C), 18.6 (d, JP-P ) 10.4,
JPt-P ) 4455, trans to O). Anal. Calcd for C63H60O2P2Pt: C, 68.40;
H, 5.47. Found: C, 67.70; H, 5.35.
Pt(OC6H4CO){(R,R)-Me-DUPHOS}; [Pt{(R,R)-Me-DUPHOS}-
(Sal)] (1d). The procedure used for the preparation of complex 1b
was followed, starting from K2PtCl4 (0.14 g, 0.33 mmol), Na2CO3 (0.10
g, 0.98 mmol), and salicylaldehyde (40 mg, 0.33 mmol), and (R,R)-
Me-DUPHOS (0.10 g, 0.33 mmol) was used instead of (R)-BINAP,
giving 0.14 g (71%) of complex 1d (recrystallized from n-hexane) as
yellow microcrystals.
Pt(OC6H4CO){(R,R)-Me-DUPHOS} (1d): IR (KBr) 1612, 1574,
1455, 1314, 1274, 882, 758, 655, 551 cm-1 1H NMR (400 MHz,
;
CDCl3) δ 7.68-7.66 (m, 2H, aromatic), 7.56-7.54 (m, 2H, aromatic),
7.30 (d, J ) 7.8, 1H, aromatic), 7.18 (dd, J ) 8.3, 7.8, 1H, aromatic),
6.91 (d, J ) 8.3, 1H, aromatic), 6.45 (dd, J ) 7.8, 7.8, 1H, aromatic),
3.55-3.25 (m, 1H, CHCH3), 3.00-2.78 (m, 2H, CHCH3), 2.60-2.20
(m, 5H, CHCH3, CH2CH2), 2.00-1.60 (m, 4H, CH2CH2), 1.49 (dd, J
) 19.5, 6.8, 3H, CHCH3), 1.27 (dd, J ) 19.1, 6.4, 3H, CHCH3), 0.85
(dd, J ) 14.7, 7.3, 3H, CHCH3), 0.84 (dd, J ) 16.6, 7.3, 3H, CHCH3);
31P NMR (160 MHz, CDCl3) δ 66.0 (d, JP-P ) 4.6, JPt-P ) 1530, trans
to C), 52.7 (d, JP-P ) 4.6, JPt-P ) 3935, trans to O). Anal. Calcd for
C25H32O2P2Pt: C, 48.31; H, 5.19. Found: C, 47.47; H, 5.10.
Pt(OC6H4CO){(R,R)-CHIRAPHOS}; [Pt{(R,R)-CHIRAPHOS}-
(Sal)] (1e). The procedure used for the preparation of complex 1b
was followed, starting from K2PtCl4 (0.97 g, 2.3 mmol), Na2CO3 (0.75
g, 7.0 mmol), and salicylaldehyde (0.29 g, 2.3 mmol), and (R,R)-
CHIRAPHOS (1.0 g, 2.3 mmol) was used instead of (R)-BINAP, giving
1.6 g (90%) of complex 1e (recrystallized from ether) as orange yellow
microcrystals.
Pt(OC6H4CO){(R,R)-CHIRAPHOS} (1e): IR (KBr) 1612, 1577,
1
1457, 1320, 1278, 1102, 881, 756, 691 cm-1; H NMR (400 MHz,
CDCl3) δ 7.95-7.75 (m, 8H, aromatic), 7.52-7.35 (m, 12H, aromatic),
7.12 (d, J ) 7.8, 1H, aromatic), 7.08 (dd, J ) 7.8, 7.8, 1H, aromatic),
6.81 (d, J ) 8.3, 1H, aromatic), 6.34 (dd, J ) 8.3, 7.8, 1H, aromatic),
2.45-2.32 (m, 1H, CHCH3), 2.30-2.15 (m, 1H, CHCH3), 1.02 (dd, J
) 11.2, 6.8, 3H, CHCH3), 0.97 (dd, J ) 13.2, 6.8, 3H, CHCH3); 31P
NMR (160 MHz, CDCl3) δ 45.7 (s, JPt-P ) 1482, trans to C), 35.9 (s,
JPt-P ) 4162, trans to O). Anal. Calcd for C35H32O2P2Pt: C, 56.68;
H, 4.35. Found: C, 55.79; H, 4.19.
A Typical Experimental Procedure for Platinum-Catalyzed
Enantioselective Aldol Reactions. (1) Catalyst Activation. To
complex 1a (26 mg, 0.025 mmol) in wet CH2Cl2 (2.2 mL, H2O 320
ppm, 0.05 mmol) was added HOTf (2.2 µL, 0.025 mmol) with a
microsyringe and stirring was continued at ambient temperature for
15 min under air.
(2) Aldol Reaction. The activated catalyst solution was cooled to
-78 °C and 2,6-lutidine (2.9 µL, 0.025 mmol) was added. To this
yellow solution was added hydrocinnamaldehyde (67 mg, 0.5 mmol),
followed by the dropwise addition of methyl trimethylsilyl dimeth-
ylketene acetal (122 mg, 0.7 mmol). The reaction atmosphere was
replaced by argon via freeze-pump-thaw (FPT) of the reaction
mixture. The reaction mixture was kept in a -25 °C freezer for 168
h at which time GC showed 96% of the starting aldehyde was
consumed. Then, the reaction mixture was poured into ice-cooled water
and extracted by CH2Cl2 (10 mL × 3). The CH2Cl2 extracts were
combined, dried over MgSO4, filtered, and concentrated in vacuo. Flash
chromatography (Wakogel (silica gel) 200, Hexane/EtOAc ) 100/0-
20/1-5/1) yielded 85 mg (0.28 mmol) of methyl 2,2-dimethyl-5-phenyl-
3-trimethylsiloxypentanoate and 46 mg (0.19 mmol) of methyl 3-hydroxy-
2,2-dimethyl-5-phenylpentanoate (desilylated product) in 94% combined
yield.
Figure 3. 31P NMR of complex A [1a was activated by HOTf (1 equiv)
in the presence of H2O (2 equiv) under air; then the atmosphere of the
NMR tube was replaced by Ar] in CD2Cl2 at 25 °C (a). KSA (10 equiv)
was introduced and 31P NMR was taken after 30 min (b) and 24 h (c).
Additional KSA (18 equiv) was introduced and 31P NMR was taken
after 4 h (d).
The products were further quantitatively deprotected by treatment
with tetrabutylammonium fluoride/THF solution (1.0 M) to give methyl
3-hydroxy-2,2-dimethyl-5-phenylpentanoate. The ee of the product was
then determined by chiral HPLC analysis (CHIRALPAK AD (Daicel
Chemical Co.)) to be 95% ee (see Supporting Information for further
details).
Pt{O(3,5-tBu2C6H2)CO}{(R)-Tol-BINAP} (1c): IR (KBr) 1625,
1439, 1246, 805 cm-1; 1H NMR (400 MHz, CDCl3) δ 7.79-7.71 (m,
4H, aromatic), 7.53 (d, J ) 8.3, 1H, aromatic), 7.47-7.17 (m, 15H,
aromatic), 7.09-7.00 (m, 4H, aromatic), 6.71 (dd, J ) 8.8, 8.8, 2H,
aromatic), 6.36 (d, J ) 7.8, 4H, aromatic), 2.39 (s, 3H, CH3C6H4),
2.35 (s, 3H, CH3C6H4), 1.95 (s, 3H, CH3C6H4), 1.94 (s, 3H, CH3C6H4),
Methyl 2,2-dimethyl-5-phenyl-3-trimethylsiloxypentanoate: IR
1
(neat) 2955, 1729, 1251, 1132, 1101, 840, 751, 699 cm-1; H NMR
(270 MHz, C6D6) δ 7.18-7.03 (m, 5H, aromatic), 4.06 (dd, J ) 8.1,

10038 J. Am. Chem. Soc., Vol. 120, No. 39, 1998
Fujimura
Figure 4. Complex B (31P NMR data) and possible structures.
Methyl 3-hydroxy-2,2-dimethyl-3-phenylpropanoate (desilylated
Scheme 5
product): IR (neat) 3454 (br), 2978, 1705, 1472, 1293, 1276, 1161,
1
1052, 771, 706 cm-1; H NMR (270 MHz, CDCl3) δ 7.25-7.22 (m,
2H, aromatic), 7.25-7.07 (m, 3H, aromatic), 4.86 (d, J ) 4.4, 1H,
CHOH), 3.30 (s, 3H, CO2CH3), 2.83 (d, J ) 4.4, OH), 1.19 (s, 3H,
C(CH3)), 1.02 (s, 3H, C(CH3)); 13C NMR (75 MHz, CDCl3) δ 177.6,
141.0, 128.1, 127.8, 127.7, 78.7, 51.5, 48.1, 22.6, 19.5; mass spectrum
(CI) m/z (rel intensity) 209 (MH+, 5), 191 (100), 102 (76), 73 (74).
Reaction Products from Cinnamaldehyde and Methyl Trimeth-
ylsilyl Dimethylketene Acetal. Methyl (E)- 2,2-dimethyl-5-phenyl-
3-trimethylsiloxy-4-pentenoate: IR (neat) 2954, 1740, 1251, 1135,
1
1066, 892, 877, 842, 745, 694 cm-1; H NMR (400 MHz, C6D6) δ
7.24-7.01 (m, 5H, aromatic), 6.51 (d, J ) 16.0, 1H, PhCH), 6.22 (dd,
J ) 16.0, 7.3, 1H, PhCHCH), 4.63 (d, J ) 7.3, 1H, CHOSiMe3), 3.39
(s, 3H, CO2CH3), 1.34 (s, 3H, C(CH3)), 1.16 (s, 3H, C(CH3)), 0.13 (s,
9H, SiMe3); 13C NMR (100 MHz, C6D6) δ 176.5, 137.1, 132.6, 128.2,
128.1, 126.8, 110.6, 79.1, 51.3, 48.7, 21.6, 20.1, 0.4; mass spectrum
(CI) m/z (rel intensity) 307 (MH+, 1), 205 (100), 157 (46), 73 (55);
HRMS, m/z calcd for C16H23O3Si (M - CH3+) 291.1427, found
291.1428.
Methyl (E)-3-hydroxy-2,2-dimethyl-5-phenyl-4-pentenoate (desi-
lylated product): IR (neat) 3483 (br), 2980, 1725, 1470, 1270, 1136,
1
1100, 970, 750, 695 cm-1; H NMR (400 MHz, CDCl3) δ 7.34-7.10
(m, 5H, aromatic), 6.58 (d, J ) 16.1, 1H, PhCH), 6.16 (dd, J ) 16.1,
6.8, 1H, PhCHCH), 4.30 (d, J ) 6.8, 1H, CHOH), 3.67 (s, 3H,
CO2CH3), 3.00-2.40 (br s, 1H, OH), 1.19 (s, 3H, C(CH3)), 1.18 (s,
3H, C(CH3)); 13C NMR (100 MHz, CDCl3) δ 177.8, 136.6, 132.9, 128.5,
127.8, 127.4,126.5, 77.8, 52.0, 47.2, 22.7, 20.0; mass spectrum (EI)
m/z (rel intensity): 234 (MH+, 2), 157 (4), 133 (100), 102 (45), 77
(20), 55 (30); HRMS, m/z calcd for C14H18O3 (M+) 234.1255, found
234.1255.
2.9, 1H, CHOSiMe3), 3.29 (s, 3H, CO2CH3), 2.88-2.78 (m, 1H,
PhCH2CHH), 2.55-2.37 (m, 1H, PhCH2CHH), 1.75-1.60 (m, 2H,
PhCH2), 1.20 (s, 3H, C(CH3)), 1.06 (s, 3H, C(CH3)), 0.13 (s, 9H,
SiMe3); 13C NMR (67.5 MHz, C6D6) δ 177.0, 142.5, 128.7, 128.6,
126.2, 77.9, 51.2, 48.4, 35.4, 33.9, 21.3, 20.9, 0.9; mass spectrum (CI)
m/z (rel intensity) 309 (MH+, 40), 219 (98), 159 (45), 117 (100), 91
(96), 73 (70); HRMS, m/z calcd for C16H25O3Si (M - CH3+) 293.1589,
found 293.1591.
Reaction Products from 3-Methylbutanal and Methyl Trimeth-
ylsilyl Dimethylketene Acetal. Methyl 2,2,5-trimethyl-3-trimeth-
ylsiloxyhexanoate: IR (neat) 2957, 1742, 1469, 1251, 1132, 1096, 1032,
1
911, 841 cm-1; H NMR (400 MHz, C6D6) δ 4.13 (dd, J ) 9.8, 2.0,
Methyl 3-hydroxy-2,2-dimethyl-5-phenylpentanoate: IR (neat)
1
3512 (br), 2951, 1723, 1455, 1275, 1135, 1076, 750, 701 cm-1; H
1H, CHOSiMe3), 3.36 (s, 3H, CO2CH3), 1.79-1.72 (m, 1H, CH3CHCH3),
1.54-1.48 (m, 1H, CH3CH(CH3)CHH), 1.26 (s, 3H, C(CH3)), 1.14 (s,
3H, C(CH3)), 1.08-1.01 (m, 1H, CH3CH(CH3)CHH), 0.89 (d, 6H, J
) 6.8, CH3CHCH3), 0.15 (s, 9H, SiMe3); 13C NMR (100 MHz, C6D6)
δ 177.0, 76.0, 51.2, 48.4, 42.7, 24.8, 24.3, 21.5, 21.4, 20.5, 0.9; mass
NMR (270 MHz, CDCl3) δ 7.31-7.19 (m, 5H, aromatic), 3.68 (s, 3H,
CO2CH3), 3.63 (m, 1H, CHOH), 3.05-2.90 (m, 1H, PhCH2CHH),
2.72-2.59 (m, 1H, PhCH2CHH), 2.58 (d, J ) 7.3, 1H, OH), 1.86-
1.52 (m, 2H, PhCH2), 1.18 (s, 3H, C(CH3)), 1.16 (s, 3H, C(CH3)); 13
C
spectrum (CI) m/z (rel intensity) 261 (MH+, 20), 245 (10), 171 (100),
NMR (75 MHz, CDCl3) δ 178.2, 142.1, 128.5, 128.4, 125.9, 76.1, 51.9,
47.1, 33.6, 32.9, 22.5, 20.4; mass spectrum (CI) m/z (rel intensity) 237
(MH+, 20), 219 (50), 159 (65), 117 (100), 91 (80), 70 (30); HRMS,
m/z calcd for C14H20O3 (M+) 236.1412, found 236.1412.
+
139 (10), 111 (12); HRMS, m/z calcd for C12H25O3Si (M - CH3
245.1576, found 245.1576.
)
Methyl 3-hydroxy-2,2,5-trimethylhexanoate (desylilated prod-
uct): IR (neat) 3512 (br), 2951, 1723, 1469, 1265, 1142, 1074 cm-1
Reaction Products from Benzaldehyde and Methyl Trimethylsilyl
Dimethylketene Acetal. Methyl 2,2-dimethyl-3-phenyl-3-trimeth-
ylsiloxypropanoate: IR (neat) 2954, 1743, 1727, 1252, 1133, 1094,
;
1H NMR (270 MHz, CDCl3) δ 3.70 (s, 3H, CO2CH3), 3.73-3.66 (m,
1H, CHOH), 2.35 (d, J ) 7.3, 1H, OH), 1.88-1.80 (m, 1H,
CH3CHCH3), 1.40-1.08 (m, 2H, CH3CH(CH3)CHH), 1.18 (s, 3H,
C(CH3)), 1.17 (s, 3H, C(CH3)), 0.95 (d, 3H, J ) 6.6, CHCH3), 0.91
(d, 3H, J ) 6.6, CHCH3); 13C NMR (67.5 MHz, CDCl3) δ 178.3, 74.7,
51.9, 47.2, 41.0, 24.9, 24.0, 22.2, 21.4, 20.4; mass spectrum (CI) m/z
(rel intensity) 189 (MH+, 40), 171 (100), 157 (28), 139 (25), 102 (97);
HRMS, m/z calcd for C10H19O2 (M - OH+) 171.1382, found 171.1382.
1
1068, 881, 843 cm-1; H NMR (270 MHz, C6D6) δ 7.28-7.27 (m,
2H, aromatic), 7.15-7.05 (m, 3H, aromatic), 5.12 (s, 1H, CHOSiMe3),
3.39 (s, 3H, CO2CH3), 1.30 (s, 3H, C(CH3)), 1.03 (s, 3H, C(CH3)),
0.00 (s, 9H, SiMe3); 13C NMR (67.5 MHz, C6D6) δ 176.6, 141.3, 128.2,
127.8, 127.6, 79.7, 51.3, 49.3, 22.0, 19.5, 0.0; mass spectrum (CI) m/z
(rel intensity) 281 (MH+, 5), 265 (10), 191 (100), 73 (90).

Pt-Catalyzed EnantioselectiVe Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 120, No. 39, 1998 10039
Reaction Products from 3,3-Dimethylbutanal and Methyl Tri-
methylsilyl Dimethylketene Acetal. Methyl 2,2,5,5-tetramethyl-3-
trimethylsiloxyhexanoate: IR (neat) 2955, 1741, 1262, 1252, 1131,
phenylpentanoate (desilylated product) in 94% combined yield together
with 54 mg of unreacted cyclohexanecarboxaldehyde (96% recovered).
The aldol products were further quantitatively deprotected by
treatment with tetrabutylammonium fluoride/THF solution (1.0 M) to
give methyl 3-hydroxy-2,2-dimethyl-5-phenylpentanoate. The ee of
the product was then determined by chiral HPLC analysis (CHIRAL-
PAK AD (Daicel Chemical Co.)) to be 94% ee.
NMR Experiments for Catalyst Activation. (a) Sample prepara-
tion was performed in a drybox under nitrogen atmosphere. A solution
of platinum complex 1a (26 mg, 0.025 mmol) in dry CD2Cl2 (1 mL)
was transferred into a dry 5-mm NMR tube. HOTf (2.2 µL, 0.025
mmol) was added with a microsyringe and thoroughly mixed. The
reaction progress was monitored by NMR.
(b) Sample preparation was performed in open air. A solution of
platinum complex 1a (26 mg, 0.025 mmol) in dry CD2Cl2 (1 mL) was
transferred into a dry 5-mm NMR tube and sealed with a rubber septa.
HOTf (2.2 µL, 0.025 mmol) was added with a microsyringe and
thoroughly mixed. The reaction progress was monitored by NMR.
(c) Sample preparation was performed in open air. A solution of
platinum complex 1a (26 mg, 0.025 mmol) in wet CD2Cl2 (1 mL) (H2O,
660 ppm, 0.05 mmol) was transferred into a 5-mm NMR tube and
sealed with a rubber septa. HOTf (2.2 µL, 0.025 mmol) was added
with a microsyringe and thoroughly mixed. The reaction progress was
monitored by NMR.
1
1098, 841 cm-1; H NMR (400 MHz, C6D6) δ 4.26 (dd, J ) 7.3, 1.5,
1H, CHOSiMe3), 3.35 (s, 3H, CO2CH3), 1.46 (dd, J ) 14.7, 7.3, 1H,
CH3C(CH3)2CHH), 1.33 (dd, J ) 14.7, 1.5, 1H, CH3C(CH3)2CHH),
1.24 (s, 3H, C(CH3)CO2CH3), 1.10 (s, 3H, C(CH3)CO2CH3), 0.93 (s,
9H, C(CH3)3), 0.19 (s, 9H, SiMe3); 13C NMR (100 MHz, C6D6) δ 177.0,
75.2, 51.2, 49.4, 47.7, 30.4, 30.2, 21.1, 21.0, 1.5; mass spectrum (CI)
m/z (rel intensity) 275 (MH+, 9), 259 (20), 185 (100), 117 (25), 73
(25); HRMS, m/z calcd for C13H27O3Si (M - CH3+) 269.1733, found
259.1733.
Methyl 3-hydroxy-2,2,5,5-tetramethylhexanoate (desilylated prod-
uct): IR (neat) 3526 (br), 2954, 1720, 1470, 1366, 1134,1074, 994
cm-1 1H NMR (270 MHz, CDCl3) δ 3.76 (dd, J ) 7.3, 7.3, 1H,
;
CHOH), 3.70 (s, 3H, CO2CH3), 1.36-1.16 (m, 2H, CCH2CH), 1.16
(s, 6H, C(CH3)2), 0.96 (s, 9H, C(CH3)3); 13C NMR (67.5 MHz, CDCl3)
δ 176.8, 72.6, 50.4, 46.5, 44.7, 28.6, 20.6, 19.1; mass spectrum (CI)
m/z (rel intensity) 204 (MH+, 45), 185 (100), 170 (12), 102 (8); HRMS,
m/z calcd for C11H21O2 (M - OH+) 185.1567, found 185.1570.
Chemoselective Platinum-Catalyzed Enantioselective Aldol Reac-
tion. To complex 1a (26 mg, 0.025 mmol) in wet CH2Cl2 (2.2 mL,
H2O 320 ppm, 0.05 mmol) was added HOTf (2.2 µL, 0.025 mmol)
with a microsyringe with stirring at ambient temperature for 15 min
under air. The activated catalyst solution was cooled to -78 °C and
2,6-lutidine (2.9 µL, 0.025 mmol) was added. To this yellow solution
were added hydrocinnamaldehyde (67 mg, 0.5 mmol) and cyclohex-
anecarboxaldehyde (56 mg, 0.5 mmol), followed by the dropwise
addition of methyl trimethylsilyl dimethylketene acetal (131 mg, 0.75
mmol). The reaction atmosphere was replaced by argon via freeze-
pump-thaw (FPT) of the reaction mixture. The reaction mixture was
kept in a -25 °C freezer for 72 h at which time GC showed 96% of
the starting hydrocinnamaldehyde was consumed while less than 3%
of the cyclohexanecarboxaldehyde was consumed. Then, the reaction
mixture was poured into ice-cooled water and extracted by CH2Cl2 (10
mL × 3). The CH2Cl2 extracts were combined, dried over MgSO4,
filtered, and concentrated in vacuo. Flash chromatography (Wakogel
(silica gel) 200, hexane/EtOAc ) 100/0-20/1-5/1) yielded 108 mg
(0.34 mmol) of methyl 2,2-dimethyl-5-phenyl-3-trimethylsiloxypen-
tanoate and 35 mg (0.14 mmol) of methyl 3-hydroxy-2,2-dimethyl-5-
Acknowledgment. The author thanks the staff of UBE
Scientific Analysis Laboratory for their assistance (especially
Mr. Tadao Gogota for NMR experiments). The author also
wishes to express his gratitude to Professor Geoffrey W. Coates
(Cornell University) for valuable discussions.
Supporting Information Available: The proof of the
enantiomer excess determination of all the reaction products
and chiral HPLC analysis conditions as well as 1H NMR spectra
for all aldol products (15 pages print/PDF). See any current
masthead page for ordering information and Web access
instructions.
JA981970D