Yb(OTf)3-TMSCl, a novel catalytic system in cross-aldol reactions

Article abstract of DOI:10.1071/CH04006

A combination of Yb(OTf)₃ and TMSCl influenced the outcome of cross-aldol reactions of cycloalkanones and benzaldehyde. Interestingly, reaction of cycloheptanone and cyclooctanone with aldehydes under the Yb(OTf)_3- TMSCl reagent system provides 3-(2-oxocycloalkyl)-3- phenylpropanals in conjunction with the corresponding aldol products.

Full text of DOI:10.1071/CH04006

Rapid Communication
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2004, 57, 655–657
Yb(OTf )3–TMSCl, a Novel Catalytic System in Cross-Aldol Reactions
Natsuko Kagawa,A Masahiro Toyota,A and Masataka IharaA,B
A Department of Organic Chemistry, Graduate School of Pharmaceutical Sciences, Tohoku University,
Aobayama, Sendai 980-8578, Japan.
B Author to whom correspondence should be addressed (e-mail: mihara@mail.pharm.tohoku.ac.jp).
A combination of Yb(OTf )3 and TMSCl influenced the outcome of cross-aldol reactions of cycloalkanones and
benzaldehyde. Interestingly, reaction of cycloheptanone and cyclooctanone with aldehydes under the Yb(OTf )3–
TMSCl reagent system provides 3-(2-oxocycloalkyl)-3-phenylpropanals in conjunction with the corresponding
aldol products.
Manuscript received: 13 January 2004.
Final version: 18 March 2004.
LA
A large variety of metals are known to act as efficient cat-
LA
alysts for the Mukaiyama aldol reaction;[1,2] however, it is
a significant problem, especially in natural product synthe-
sis, to generate silyl enol ethers from highly functionalized
and complex ketones. Metals with appropriate Lewis acid-
ity and oxophilicity which could be useful in direct catalytic
aldol reactions[3] are very few, e.g. Zn and Mg. For instance,
Shibasaki[4] and Trost[5] and their coworkers have indepen-
dently reported direct catalytic asymmetric aldol reactions
using Zn. Furthermore, Evans[6] and his colleagues have
recently demonstrated Mg-catalyzed direct aldol reactions.
In general, strong Lewis acidity promotes rapid forma-
tion of a metal enolate (A) and effects the 1,2-addition
to an aldehyde (B). The formation of a quite stable bond
between oxygen and a metal such as B and Ti makes it
difficult for the metal to dissociate easily and be involved
in the catalytic cycle again (C). It is believed that this is
a crucial step in realizing a direct catalytic aldol reaction
(Scheme 1).
O
O
O
O
O
OH
RCHO
(B)
ꢀ
LA
R
R
(A)
(C)
Scheme 1.
by 1H NMR spectroscopy, after transformation into the cor-
responding β-hydroxy ketones.
While aldol reaction of cyclopentanone and benzaldehyde,
without TMSCl, gave the desired aldol product in only 6%,
the addition of TMSCl increased the combined yield of aldol
products to 23% (run 1). However, when the reaction was
performed with cyclohexanone, we observed an almost four-
fold increase in the yield of aldol products (run 2). When the
same reaction was carried out in the absence of benzalde-
hyde, 1-trimethylsilyloxycyclohex-1-ene[12] was formed. In
addition, conducting the reaction in the absence ofYb(OTf )3
did not provide the silyl enol ether at all, even after 48 h of
stirring. We suppose that both the ketone I and aldehyde III
could be activated as shown in Scheme 2, and silyl enol ether
II would attack III to yield IV, although we cannot leave out
the other probability, namely the activation of the carbonyl
group by an intermediate of type V.[13]
In order to get some evidence, we arrived at a correlation
between the amount of TMSCl and the product yield. The
results are depicted graphically in Fig. 1. It was found that
yields increased linearly with increasing amounts of TMSCl,
probably because two equivalents of TMSCl might be con-
sumed in order to activate both carbonyl groups, and one
equivalent of TMSCl will be further used to form the silyl
enol ether of cyclohexanone, which could be regenerated
by a retro-aldol reaction of the partially formed Yb aldolate
(Scheme 1). This speculation was supported by the ratio of
β-hydroxy ketone to the total amount of aldol products
formed, as shown in Fig. 1; namely, compared to 100 mol% of
In the course of our synthetic studies on mycalamide,[7] we
recently observed an improved Mukaiyama aldol reaction[8]
using a combination of Yb(OTf )3 and TMSCl (trimethyl-
silyl chloride) to provide the desired aldol product. This
combination was adapted from the conditions originally
developed for an imino ene reaction by Nakagawa and her
coworkers.[9,10]
Webecameinterestedinexaminingthereactionconditions
for cross-aldol reactions. In this paper, the first example of a
Yb(OTf )3-catalyzed cross-aldol reaction via a silyl enol ether
is described.[11]
Several examples of the Yb(OTf )3-catalyzed aldol reac-
tion of cyclic ketones with aldehydes are shown in Table 1.
All reactions were carried out at room temperature using 1.0
equivalent of aldehyde, 3.0 equivalents of TMSCl, and 3.8
equivalents of Et3N in the presence of 10 mol% Yb(OTf )3.
The diastereomeric ratios of aldol products were determined
© CSIRO 2004
10.1071/CH04006
0004-9425/04/070655

656
N. Kagawa, M. Toyota and M. Ihara
A
Table 1. Cross-aldol reactions using a combination of Yb(OTf )3 and TMSCl
O
O
OTMS
O
OH
O
R
CHO
RCHO
R
R
ꢀ
ꢀ
( )n
( )n
( )n
( )n
Yb(OTf)3, TMSCl,
NEt3, CH2Cl2, room temp.
1
2
3
4
Run
no.
Ketone
used
Aldehyde
used
Time
[h]
Yield 3 [%]B
(without TMSCl)
Yield [%]B
3
2
4
1
2
3
4
Cyclopentanone
Cyclohexanone
Cyclohexanone
Cyclohexanone
PhCHO
PhCHO
p-MeOPhCHO
36
37
40
48
6 (59 : 41)
2 (53 : 47)
3 (65 : 35)
4 (54 : 46)
18 (48 : 52)
70 (64 : 36)
71 (60 : 40)
33 (52 : 48)
5 (50 : 50)
2 (64 : 36)
0
0
0
0
0
CHO
3 (61 : 39)
Ph
5
6
Cycloheptanone
Cyclooctanone
PhCHO
PhCHO
36
42
3 (33 : 67)
0
43 (62 : 38)
62 (99 : 31)
0
0
5 (87 : 13)C
5 (69 : 31)C
A Yb(OTf )3 was obtained from Aldrich.
diastereomer was not determined.
B Anti-to-syn ratios are shown in parentheses.
C The relative configuration of each isolated
Yb(OTf)3
Si Cl
Yb(OTf)3
Si Cl
ꢀN
Si
Et3N
Si Cl Yb(OTf)3
OTMS
O
O
H
NEt3
H
Me3SiH
H
Et3N
A
B
I
II
III
Si Cl
Yb(OTf)3
HCl
H
Nꢀ
N
O
OTf
TfO
O
O
OTMS
NEt3
SiMe3
Yb Cl
H
Ph
OTf
C
F
D
E
R1
R2
TMSO
Nꢀ
IV
V
OTMS
Ph
Nꢀ
H
Scheme 2. Proposed reaction mechanism of the Yb(OTf )3-catalyzed
aldol reaction.
TMSOH
Ph
G
O
O
OR
Ph
R ꢁ H or TMS
R1
O
O
Ph
H3Oꢀ
CHO
N
ꢀ
PhCHO
R2
Ph
R1
R2
H
H
I
0.4
0.3
0.2
0.1
70
60
50
40
30
20
10
Scheme 3. Plausible mechanism for the generation of propanal
derivatives.
The substituent effects for this reaction were also investi-
gated. An electron-donating group on the aromatic ring in
run 3 had little effect on the aldol reaction. Further, the
use of trans-cinnamaldehyde resulted in a decreased yield
(run 4). Surprisingly, when the reaction was performed with
cycloheptanone and cyclooctanone, 3-(2-oxocycloalkyl)-3-
phenylpropanals 4 were isolated in 5% yield, together with
aldol products in good-to-moderate yield (runs 5 and 6). In
these cases, we speculate that an activated benzaldehyde E is
expected to react with enamine D, generated from triethyl-
amine, TMSCl, and Yb(OTf )3, to provide, via F, iminium
salt G,[14] which was attacked by the silyl enol ether of the
ketone at the less hindered position. Finally, hydrolysis of H
takes place leading to the formation of propanal derivative I
(Scheme 3).
100
200
300
TMSCl (mol%)
Fig. 1. Reaction profile of the cross-aldol reaction of cyclohex-
anone with benzaldehyde in the presence of 10 mol% Yb(OTf )3 and
100–300 mol%TMSCl.Theopentrianglereferstotheratioofβ-hydroxy
ketone to aldol products.
TMSCl, there was a sharp decrease in the ratio from 0.4 to
0.1 when 200 mol% of TMSCl was used. Further, complete
consumption of starting materials with 300 mol% of TMSCl
clearly indicates that this amount of TMSCl is essential for
the activation of both carbonyl groups as well as improvement
of the yield.
This theory is supported by the following control exper-
iment. When benzaldehyde was subjected to the reaction

A Novel Catalytic System in Cross-Aldol Reactions
657
conditions, only cinnamaldehydes (trans-to-cis ratio 59 : 41)
were obtained in 35% yield. The aldol reactions of Yb eno-
lates are believed to proceed by way of a Zimmerman–Traxler
transition state in which the metal atom is coordinated tightly
to the oxygen atoms of the aldehyde to provide preferentially
anti aldol products.[15] Since anti aldol products were not
observed in very high ratio under our reaction conditions, we
speculate that the present aldol reactions proceed through
silyl enol ether intermediates. Indeed, TLC analysis supports
the formation of silyl enol ethers of the cycloalkanones in
every case. In addition, attempts to conduct the present aldol
reaction in the absence of Et3N met with no success. In the
case of acyclic ketones, the aldol products were not formed
in reasonable yields.
In conclusion, we have demonstrated the cross-aldol reac-
tion of cycloalkanones with aldehydes using 10 mol% of
Yb(OTf )3 in the presence of TMSCl and Et3N. The function
of TMSCl in this reaction is unclear; however, we assume
that TMSCl facilitates enolate formation and activates the
carbonyl oxygen of aldehydes. Interestingly, reactions of
both cycloheptanone and cyclooctanone with benzaldehyde
under the Yb(OTf )3–TMSCl reagent system provide 3-(2-
oxocycloalkyl)-3-phenylpropanals in conjunction with the
corresponding aldol products.
(b) B. List, Tetrahedron 2002, 58, 5573. doi:10.1016/S0040-
4020(02)00516-1
(c) M. Shibasaki, N. Yoshikawa, Chem. Rev. 2002, 102, 2187.
doi:10.1021/CR010297Z
[4] (a) N. Yoshikawa, N. Kumagai, S. Matsunaga, G. Moll,
T. Ohshima, T. Suzuki, M. Shibasaki, J. Am. Chem. Soc.
2001, 123, 2466. doi:10.1021/JA015580U
(b) N. Kumagai, S. Matsunaga, T. Kinoshita, S. Harada,
S. Okada, S. Sakamoto, K. Yamaguchi, M. Shibasaki, J. Am.
Chem. Soc. 2003, 125, 2169. doi:10.1021/JA028926P
[5] (a) B. M. Trost, H. Ito, E. R. Silcoff, J. Am. Chem. Soc. 2001,
123, 3367. doi:10.1021/JA003871H
(b) B. M. Trost, E. R. Silcoff, H. Ito, Org. Lett. 2001, 3, 2497.
doi:10.1021/OL0161211
[6] (a) D. A. Evans, J. S. Tedrow, J. T. Shaw, W. Downey, J. Am.
Chem. Soc. 2002, 124, 392. doi:10.1021/JA0119548
(b) D. A. Evans, W. Downey, J. T. Shaw, J. S. Tedrow, Org. Lett.
2002, 4, 1127. doi:10.1021/OL025553O
[7] (a) M. Toyota, M. Hirota, H. Hirano, M. Ihara, Org. Lett. 2000,
2, 2031. doi:10.1021/OL000072V
(b) M. Toyota, M. Hirota, H. Hirano, K. Natsuko, M. Ihara, 18th
International Congress of Heterocyclic Chemistry 2001, p. 295
(International Society of Heterocyclic Chemistry: Yokohama).
[8] For lanthanide metal catalyzed Mukaiyama aldol reactions, see:
(a) K. Takai, C. H. Heathcock, J. Org. Chem. 1985, 50, 3247.
(b) A. E. Vougioukas, H. B. Kagan, Tetrahedron Lett. 1987, 28,
5513. doi:10.1016/S0040-4039(00)96767-4
(c) L. Gong, A. Streitwieser, J. Org. Chem. 1990, 55, 6235.
(d) K. Mikami, M. Terada, T. J. Nakai, J. Chem. Soc., Chem.
Commun. 1993, 343.
(e) S. Kobayashi, I. Hachiya, H. Ishitani, M. Araki, Synlett
1993, 472. doi:10.1055/S-1993-22495
Experimental
Representative Procedure
(f) N. Giuseppone, P. V. Weghe, M. Mellah, J. Collin, Tetra-
hedron 1998, 54, 13129. doi:10.1016/S0040-4020(98)00791-1
[9] (a) M. Yamanaka, A. Nishida, M. Nakagawa, Org. Lett. 2000,
2, 159. doi:10.1021/OL991260S
(b) M. Yamanaka, A. Nishida, M. Nakagawa, J. Org. Chem.
2003, 68, 3112. doi:10.1021/JO0268153
[10] For other combinations of Lewis acid–TMSX reagents, see:
(a) S. Danishefsky, T. Kitahara, J. Am. Chem. Soc. 1974,
96, 7807.
To a suspension of Yb(OTf )3 (60.0 mg, 9.65 × 10−2 mmol) and
chlorotrimethylsilane (3.70 × 10−1 mL, 2.90 mmol) in CH2Cl2 (8 mL)
was added a solution of cyclohexanone (9.60 × 10−2 mL, 9.65 ×
10−1 mmol) in CH2Cl2 (2 mL) followed by triethylamine (5.00 ×
10−1 mL, 3.62 mmol). After 40 min, benzaldehyde (1.00 × 10−1 mL,
9.80 × 10−1 mmol) was added and the resulting solution was stirred
for 37 h. A solution of NaHCO3 (10 mL, saturated) was added, and
the resulting mixture was poured into EtOAc (20 mL). The layers
were separated and the aqueous layer was further extracted with
EtOAc (2 × 20 mL). The combined organic extracts were washed with
brine (30 mL), dried (MgSO4), and concentrated to provide a brown
oil. Purification of the crude product by flash chromatography (1%
EtOAc/hexane) provided the silyl aldolate (1.87 × 10−1 g, 70%) and
the aldol (4.7 mg, 2%) as colorless oils.
(b) Y. Yoshida, N. Matsumoto, R. Hamasaki, Y. Tanabe, Tetra-
hedron Lett. 1999, 40, 4227. doi:10.1016/S0040-4039(99)
00682-6
(c) A. Kakuuchi, T. Taguchi, Y. Hanzawa, Tetrahedron Lett.
2001, 42, 1547. doi:10.1016/S0040-4039(00)02282-6
(d) A. Dilman, S. L. Ioffe, Chem. Rev. 2003, 103, 733, and
references therein. doi:10.1021/CR020003P
References
See also references [6a] and [8c].
[11] S. Kobayashi, M. Sugiura, H. Kitagawa, W. W. Lam, Chem.
Rev. 2002, 102, 2227. doi:10.1021/CR010289I
[12] Y. Ito, S. Fujii, M. Nakatsuka, F. Kawamoto, T. Saegusa, Org.
Synth., Collect. Vol. 6 1988, 327.
[13] Y. Onishi, T. Ito, M. Yasuda, A. Baba, Tetrahedron 2002, 58,
8227. doi:10.1016/S0040-4020(02)00972-9
[14] (a) P. Bharathi, M. Periasamy, Org. Lett. 1999, 1, 857.
doi:10.1021/OL990745D
(b) R. N. Ram, I. J. Charles, Chem. Res. (S) 2000, 540.
[15] S. Fukuzawa, T. Tsuchimoto, T. Kanai, Bull. Chem. Soc. Jpn
1994, 67, 2227.
[1] For reviews of catalytic Mukaiyama aldol reactions, see:
(a) T. Mukaiyama, Org. React. 1982, 28, 203.
(b) S. G. Nelson, Tetrahedron: Asymmetry 1998, 9, 357. doi:
10.1016/S0957-4166(97)00634-4
(c) N. Giuseppone, P. V. Weghe, M. Mellah, J. Collin, Tetra-
hedron 1998, 54, 13129. doi:10.1016/S0040-4020(98)00791-1
(d) R. Mahrwald, Chem. Rev. 1999, 99, 1095. doi:10.1021/
CR980415R
[2] For a recent review on chiral Lewis base catalyzed aldol
reactions, see: S. E. Denmark, R. A. Stavenger, Acc. Chem.
Res. 2000, 33, 432. doi:10.1021/AR960027G
[3] For reviews of direct catalytic aldol reactions, see:
(a) B. Alcaide, P. Almendros, Eur. J. Org. Chem. 2002,
1595.
doi:10.1002/1099-0690(200205)2002:10<1595::AID-
EJOC1595>3.0.CO;2-M