Mechanistic investigations of the Mukaiyama aldol reaction as a two part enantioselective reaction

Article abstract of DOI:10.1016/j.tetlet.2008.12.083

Herein, we report a mechanistic investigation of an enantioselective tandem Mukaiyama aldol reaction, consisting of a carbon-carbon bond-forming reaction and a silylation protection step in which the enantioselectivity results exclusively from the silylation step. The reaction is carried out in the presence of a Lewis base paired with a chiral quarternary ammonium salt. Mechanistic studies indicate that the enantioselectivity of the silylation step is a kinetic resolution of the aldolate intermediate. The effects of sterics and electronics on the aldehyde starting material are also presented.

Full text of DOI:10.1016/j.tetlet.2008.12.083

Tetrahedron Letters 50 (2009) 1164–1166
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Mechanistic investigations of the Mukaiyama aldol reaction as a two part
enantioselective reaction
*
Sachin G. Patel, Sheryl L. Wiskur
Department of Chemistry and Biochemistry, University of South Carolina, 631 Sumter St., Columbia, SC 29208, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein, we report a mechanistic investigation of an enantioselective tandem Mukaiyama aldol reaction,
consisting of a carbon–carbon bond-forming reaction and a silylation protection step in which the enanti-
oselectivity results exclusively from the silylation step. The reaction is carried out in the presence of a
Lewis base paired with a chiral quarternary ammonium salt. Mechanistic studies indicate that the enanti-
oselectivity of the silylation step is a kinetic resolution of the aldolate intermediate. The effects of sterics
and electronics on the aldehyde starting material are also presented.
Received 18 September 2008
Revised 11 December 2008
Accepted 14 December 2008
Available online 24 December 2008
Ó 2008 Elsevier Ltd. All rights reserved.
The Mukaiyama aldol reaction (Eq. 1) is a useful and powerful
reaction for the synthesis of complex products.1 Its utility arises
from the ability of the reaction to form a carbon–carbon bond
and generate up to two stereocenters under mild conditions.2
While this reaction has been intensely examined, most researchers
have overlooked the fact that this transformation is actually a com-
bination of two reactions: a carbon–carbon bond-forming reaction
and a protection step (silylation of the newly formed alkoxide) (Eq.
2). While in some reactions this happens simultaneously,3 there
are other examples where this is a stepwise process4 allowing for
an asymmetric silylation to occur. Most researchers have focused
on the enantioselectivity of the carbon–carbon bond-forming step,
while never realizing that the silylation step can also be enantiose-
lective. We have identified reaction conditions where the enanti-
oselectivity of our Mukaiyama aldol reaction arises exclusively
from the silylation step. In this process, the first step is not enan-
tioselective, yielding racemic aldolate, which is then enantioselec-
tively silylated in the second step. Thus, this is an example of a
kinetic resolution employing a silyl protecting group.5 In this
study, we have explored the mechanism of this tandem reaction
as well as the effect of sterics and electronics with regard to the
aldehyde starting material.
While searching for a new asymmetric Lewis base-catalyzed
Mukaiyama aldol reaction,6 it was discovered that more than just
an aldol reaction was taking place. Specifically, the investigation
began with the addition of a TMS silyl ketene acetal to benzalde-
hyde catalyzed by a Lewis base chiral ammonium salt, a methyl-
ated cinchona alkaloid paired with acetate6e,7 (CD-Me+AcOꢀ). The
reaction produced the expected products, a combination of the
alcohol and silylated product, which through a traditional TBAF
workup yielded the alcohol 1 (Eq. 3). We were initially disap-
pointed to find that the alcohol was racemic. However, upon exam-
ination of the two products (1 and 2) prior to the TBAF
deprotection (Eq. 4), it was discovered that the alcohol product
(1) was enantioenriched and the silylated product (2) was enantio-
enriched in the opposite enantiomer. What we believe to be hap-
ð3Þ
ð4Þ
ð1Þ
ð2Þ
* Corresponding author. Tel.: +1 803 777 8143; fax: +1 803 777 9521.
E-mail address: wiskur@mail.chem.sc.edu (S.L. Wiskur).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.12.083

S. G. Patel, S. L. Wiskur / Tetrahedron Letters 50 (2009) 1164–1166
1165
These salts become the new Lewis base catalyst, where the alkox-
ide activates the next silyl ketene acetal for carbon–carbon bond
formation.11 The enantioselectivity of the reaction arises from the
different reactivities of the diastereomeric salts with the silyl
source (II). While the silyl source could be the silyl ketene acetal
or the silyl acetate,12 it is more likely to be the silyl ketene acetal
since it is present in higher concentrations (5:1). One alkoxide
enantiomer becomes silylated, while the other remains as the salt
resulting in the alcohol product upon protonation. With each cat-
alytic turnover, the salt becomes more enantiomerically enriched
resulting in the separation of the alkoxide enantiomers.
In order to investigate the mechanism, an experiment was per-
formed to show that the diastereomeric salts could generate enan-
tiomeric excess when introduced as the catalyst at the start of the
reaction. Mukaiyama has shown that lithium alkoxide can catalyze
the reaction11, so the next step was to pair the aldolate with a chi-
ral quarternary ammonium cation. The diastereomeric salts were
preformed and subjected to a mixture of silyl ketene acetal and
aldehyde. The reaction progressed as expected forming products
1 and 2, and an ee of 63% was determined for 1 (Eq. 5). This showed
that the diastereomeric salts could catalyze carbon–carbon bond
formation while selectively silylating one enantiomer of the alkox-
ide resulting in a kinetic resolution.
Figure 1. A plot of conversion versus enantiomeric excess showing that the ee of 1
changes over the course of the reaction.
pening is a kinetic resolution via an enantioselective silylation of
the racemic alkoxide intermediate generated, in other words an
enantioselective separation of the two alkoxide enantiomers
through derivatization of one of the enantiomers. This process
was further studied to verify that the enantioselectivity was arising
from the second step and not from the first bond-forming step.
A kinetic resolution is a separation of enantiomers through selec-
tively reacting one enantiomer over the other in an asymmetric
reaction. When the reaction involves a simple enantioselective
derivatization, without altering the stereocenter, the starting
material and product should be enriched in the opposite enantio-
mers. The degree of enrichment depends on the conversion.8 The
products from our system were enriched in the opposite enantio-
mers, the alcohol product 68% (S)9 and the deprotected silylated
product 8% (R) (Eq. 4). Thus, the low ee of the deprotected product
and a high ee of the remaining alkoxide ‘starting material’ indi-
cated that the reaction proceeded to a high conversion, which is
consistent with the observed ratio of silylated to unsilylated prod-
uct (6:1). To further confirm that this process is a kinetic resolu-
tion, the ee of the reaction was examined at different degrees of
conversion. Figure 1 shows that enantioselectivity of the alcohol
varies over the course of the reaction and increases with increasing
conversion which is consistent with a kinetic resolution process.
Scheme 1 shows the proposed mechanism. The acetate Lewis
base10 activates the silyl ketene acetal for nucleophilic attack on
the aldehyde yielding the racemic alkoxide which is paired with
the chiral cinchonidine cation forming diastereomeric salts (I).
ð5Þ
Next, the effect of changing the sterics and electronics of the aro-
matic aldehyde was explored. It was discovered that the electronics
in the para position (electron-donating and -withdrawing) did not
have a significant effect on the enantioselectivity of the reaction
or selectivity factor13 (Table 1, entries f–h) compared to benzalde-
hyde (entry a). Since the aromatic ring is not conjugated with the
nucleophilic alkoxide, the electronics are presumably too far away
to have a large effect on the selectivity. It was also discovered that
by placing a methyl group in the ortho position (entry e), the
enantioselectivity was not altered, again showing that sterics in
the ortho position were not affecting the reaction. The one factor
that did have an effect on the enantioselectivity was the presence
of a halogen in the ortho position. The enantioselectivity jumped
Table 1
The results of the Mukaiyama aldol kinetic resolution with different aldehydes.
Entry
R
Ratio (1:2)
Yield (1 + 2)a
ee of 1 (%)
s
a
b
c
d
e
f
H
2-F
1:6
1:3.5
1:4
1:7
1:7
1:6
1:5
1:13
85
78
81
87
88
86
83
93
68
84
84
86
72
72
78
69
2.1
3.7
3.4
2.8
2.2
2.3
2.7
1.8
2-Cl
2-Br
2-CH3
4-F
4-Br
4-OCH3
g
h
a
Yields were determined by 1H NMR using 2,2,6,6-tetramethylbenzo-[1,2-d;4,5-
Scheme 1. Proposed mechanism for the enantioselective silylation of the alkoxide
intermediate.
d0]bis[1,3]dioxole as an internal standard. Results are an average of multiple runs.

1166
S. G. Patel, S. L. Wiskur / Tetrahedron Letters 50 (2009) 1164–1166
R.; Addie, M. S.; Crawforth, C. M.; Dale, J. W.; Franci, X.; Pizzonero, M.; Donald,
C.; Taylor, R. J. K. Tetrahedron 2008, 64, 4778–4791; (c) Zanardi, F.; Curti, C.;
Sartori, A.; Rassu, G.; Roggio, A.; Battistini, L.; Burreddu, P.; Pinna, L.; Pelosi, G.;
Casiraghi, G. Eur. J. Org. Chem. 2008, 2273–2287; (d) Li, H.; Loh, T.-P. J. Am.
Chem. Soc. 2008, 130, 7194–7195.
from the high 60’s with benzaldehyde to the mid 80’s with F, Cl, and
Br (entries b–d), and there was an improvement in the selectivity
factors as well. This selectivity could be the result of the partial neg-
ative charge of the halogens aiding in a stronger association with
the cationic catalyst, slowing down the alkoxide’s reactivity.
In conclusion, we have broken apart the Mukaiyama aldol reac-
tion to show that enantioselectivity can arise from the protection
step and not from the carbon–carbon bond-forming step. The
enantioselectivity is a result of a kinetic resolution of the racemic
alkoxide intermediates through enantioselective silylation. While
this system did not produce high selectivities, it is interesting to
note that there is the possibility of this phenomenon occurring in
other similar systems resulting in the chance of two enantioselec-
tive competing processes. This could lead to false results if crude
reaction mixtures are screened, and might warrant further investi-
gation. We ultimately were inspired by this system to explore a
more direct kinetic resolution employing asymmetric silylation
reactions where further efforts will be focused.
3. (a) Wong, C. T.; Wong, M. W. J. Org. Chem. 2007, 72, 1425–1430; (b) Denmark, S.
E.; Pham, S. M.; Stavenger, R. A.; Su, X.; Wong, K.-T.; Nishigaichi, Y. J. Org. Chem.
2006, 71, 3904–3922; (c) Myers, A. G.; Kephart, S. E.; Chen, H. J. Am. Chem. Soc.
1992, 114, 7923–7924.
4. (a) Denmark, S. E.; Chen, C.-T. Tetrahedron Lett. 1994, 35, 4327–4330; (b)
Carreira, E. M.; Singer, R. A. Tetrahedron Lett. 1994, 35, 4323–4326.
5. (a) Issenhuth, J. T.; Dagorne, S.; Bellemin-Laponnaz, S. J. Mol. Catal. A 2008,
286, 6–10; (b) Karatas, B.; Rendler, S.; Fröhlich, R.; Oestreich, M. Org.
Biomol. Chem. 2008, 6, 1435–1440; (c) Klare, H. F. T.; Oestreich, M. Angew.
Chem., Int. Ed. 2007, 46, 9335–9338; (d) Zhao, Y.; Mitra, A. W.; Hoveyda,
A. H.; Snapper, M. L. Angew. Chem., Int. Ed. 2007, 46, 8471–8474; (e) Zhao,
Y.; Rodrigo, J.; Hoveyda, A. H.; Snapper, M. L. Nature 2006, 443, 67–70; (f)
Isobe, T.; Fukuda, K.; Araki, Y.; Ishikawa, T. Chem. Commun. 2001, 243–
244.
6. (a) Song, J. J.; Tan, Z.; Reeves, J. T.; Yee, N. K.; Senanayake, C. H. Org. Lett. 2007, 9,
1013–1016; (b) Denmark, S. E.; Fan, Y.; Eastgate, M. D. J. Org. Chem. 2005, 70,
5232–5248; (c) Shen, Z.-L.; Ji, S.-J.; Loh, T.-P. Tetrahedron Lett. 2005, 46, 507–
508; (d) Nakajima, M.; Orito, Y.; Ishizuka, T.; Hashimoto, S. Org. Lett. 2004, 6,
3763–3765; (e) Mukaiyama, T.; Fujisawa, H.; Nakagawa, T. Helv. Chim. Acta
2002, 85, 4518–4531; (f) Denmark, S. E.; Stavenger, R. A. J. Am. Chem. Soc. 2000,
122, 8837–8847; (g) Matsukawa, S.; Okano, N.; Imamoto, T. Tetrahedron Lett.
2000, 41, 103–107; (h) Shioiri, T.; Bohsako, A.; Ando, A. Heterocycles 1996, 42,
93–97; (i) Ando, A.; Miura, T.; Tatematsu, T.; Shioiri, T. Tetrahedron Lett. 1993,
34, 1507–1510.
Acknowledgments
We gratefully acknowledge support from the American Chemi-
cal Society—Petroleum Research Fund (44683-G1), the University
of South Carolina Research and Productive Scholar Award, and start
up funds from the University of South Carolina. We would also like
to thank Will Tilford for the internal standard he supplied, and Ken
Shimizu for helpful discussions.
7. (a) Nakagawa, T.; Fujisawa, H.; Nagata, Y.; Mukaiyama, T. Bull. Chem. Soc. Jpn.
2004, 77, 1555–1567; (b) Nakagawa, T.; Fujisawa, H.; Mukaiyama, T. Chem. Lett.
2003, 32, 462–463; (c) Nakagawa, T.; Fujisawa, H.; Mukaiyama, T. Chem. Lett.
2003, 32, 696–697.
8. (a) Keith, J. M.; Larrow, J. F.; Jacobsen, E. N. Adv. Synth. Catal. 2001, 343, 5–26;
(b) Vedejs, E.; Jure, M. Angew. Chem., Int. Ed. 2005, 44, 3974–4001.
9. Kobayashi, S.; Ishitani, H.; Yamashita, Y.; Ueno, M.; Shimizu, H. Tetrahedron
2001, 57, 861–866.
References and notes
10. Counterions such as iodide and chloride cannot initiate the reaction, therefore
starting material is recovered.
11. Fujisawa, H.; Nakagawa, T.; Mukaiyama, T. Adv. Synth. Catal. 2004, 346, 1241–
1246.
12. If the silyl acetate delivers the silyl group, the alkoxide should still be
enantioselectively silylated and the initial catalyst will be regenerated, which
will start the catalytic cycle over again.
1. For reviews see: (a) Carreira, E. M.. In Comprehensive Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer: Heildelberg, 1999; Vol.
3, pp 997–1065; (b) Nelson, S. G. Tetrahedron: Asymmetry 1999, 9, 357–389; (c)
Palomo, C.; Oiarbide, M.; García, J. M. Chem. Eur. J. 2002, 8, 36–44; (d) Gröger,
H.; Vogl, E. M.; Shibasaki, M. Chem. Eur. J. 1998, 4, 1137–1141.
13. Selectivity factor (s) = (rate of fast acting enantiomer)/(rate of slow acting
enantiomer). See Ref. 8.
2. For recent examples of research employing the Mukaiyama Aldol reaction see:
(a) Yang, J.-H.; Liu, J.; Hsung, R. P. Org. Lett. 2008, 10, 2525–2528; (b) Webb, M.