Lewis base activation of Lewis acids: Catalytic, enantioselective addition of glycolate-derived silyl ketene acetals to aldehydes

Article abstract of DOI:10.1021/jo8006539

(Chemical Equation Presented) A catalytic system involving silicon tetrachloride and a chiral, Lewis basic bisphosphoramide catalyst is effective for the addition of glycolate-derived silyl ketene acetals to aldehydes. It was found that the sense of diast

Full text of DOI:10.1021/jo8006539

Lewis Base Activation of Lewis Acids: Catalytic, Enantioselective
Addition of Glycolate-Derived Silyl Ketene Acetals to Aldehydes
Scott E. Denmark* and Won-jin Chung
Department of Chemistry, Roger Adams Laboratory, UniVersity of Illinois, Urbana, Illinois 61801
denmark@scs.uiuc.edu
ReceiVed March 22, 2008
A catalytic system involving silicon tetrachloride and a chiral, Lewis basic bisphosphoramide catalyst is effective
for the addition of glycolate-derived silyl ketene acetals to aldehydes. It was found that the sense of
diastereoselectiVity could be modulated by changing the size of the substituents on the silyl ketene acetals. In
general, the trimethylsilyl ketene acetals derived from methyl glycolates with a large protecting group on the
R-oxygen provide enantiomerically enriched R,ꢀ-dihydroxy esters with high syn-diastereoselectivity, whereas
the tert-butyldimethylsilyl ketene acetals derived from bulky esters of R-methoxyacetic acid provide
enantiomerically enriched R,ꢀ-dihydroxy esters with high anti-diastereoselecitvity.
Introduction
ties, it is necessary to control the geometry of the starting olefins
prior to the asymmetric dihydroxylations or epoxidation since
these methods are highly stereospecific.
Stereodefined 1,2-diol units are found in a number of natural
products such as macrolides and carbohydrates as well as in
many important organic compounds such as chiral ligands in
asymmetric catalysis. 1,2-Diols are also very versatile precursors
that can be easily transformed to a variety of useful structures.
There are several different approaches that can be envisioned
for the preparation of 1,2-diols (Scheme 1): (1) sequential
reduction of 1,2-diketones, (2) oxidation of alkenes via either
direct dihydroxylations or epoxidation and opening, and (3)
carbon-carbon bond-forming reaction between the two oxygen
bearing carbons. Among these approaches, OsO4-catalyzed
enantioselective dihydroxylation of olefins developed by Sharp-
less et al. is widely recognized as one of the most practical
methods for syntheses of enantiomerically pure 1,2-diols.1
Although syn-1,2-diol can be prepared with high enantioselec-
tivities by this method, the synthesis of enantiomerically pure
anti-1,2-diols suffers from low stereoselectivity. Moreover, the
resulting two hydroxyl groups are not differentiated. More
efficient synthesis of anti-1,2-diols can be achieved by asym-
metric epoxidation and ring opening.2 To achieve high selectivi-
SCHEME 1
Background
The aldol reaction of glycolate esters with aldehydes repre-
sents a third alternative to the preparation of stereodefined 1,2-
(1) (a) Wang, L.; Sharpless, K. B. J. Am. Chem. Soc. 1992, 114, 7568–
7570. (b) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483–2547.
(2) (a) Shi, Y. Acc. Chem. Res. 2004, 37, 488–496. (b) Yang, D. Acc. Chem.
Res. 2004, 37, 497–505. (c) Xia, Q.-H.; Ge, H.-Q.; Ye, C.-P.; Liu, Z.-M.; Su,
K.-X. Chem. ReV. 2005, 105, 1603–1662.
4582 J. Org. Chem. 2008, 73, 4582–4595
10.1021/jo8006539 CCC: $40.75 2008 American Chemical Society
Published on Web 05/28/2008

Lewis Base ActiVation of Lewis Acids
diol units (Scheme 2). In this approach, a number of well-
developed platforms of the aldol reaction allow for the selective
formation of both syn- and anti-diastereomers.3 Although the
level of diastereoselectivity is often limited by the geometrical
purity of enolate, the highly syn-selective enolization of glycolate
esters alleviates this problem. Also, in the case of Mukaiyama-
type aldol reactions,4 it is often observed that both E and Z
isomers of enol ethers afford the same diastereomers (diaste-
reoconvergent) so that the requirement of the selective formation
of geometrically homogeneous enol ethers is unnecessary. In
addition, the two hydroxyl groups of the glycolate aldol product
are intrinsically differentiated.
with high selectivity, an excess of boron triflate and amine bases
were necessary. Moreover, additional steps were required for
the removal and recovery of the auxiliary.
2. Catalytic, Enantioselective Glycolate Aldol Reactions.
In contrast to the successful glycolate aldol additions using chiral
auxiliaries, only a few studies of catalytic, enantioselective
glycolate aldol reactions have been reported. Kobayashi and
co-workers have developed a catalytic system that involves a
tin-based Lewis acid and chiral amines for the asymmetric
glycolate aldol reaction.8 Both diastereomers can be obtained
in high diastereoselectivity and enantioselectivity by changing
both the geometry of ketene acetals and the structure of chiral
amines. However, this method has some disadvantages that limit
its practicality: (1) an excess of tin reagents is employed and a
stoichiometric amount of the chiral amine is required in some
cases, (2) tin reagents had to be added slowly for several hours
via syringe pump, and (3) aliphatic aldehydes are problematic
substrates. Therefore, the development of a new catalytic system
that requires only a substoichiometric amount of a chiral source
and that is applicable to a wide range of aldehydes is highly
desirable.
3. Lewis Base Activation of Lewis Acids. The generation
of chiral Lewis acid catalysts usually involves the complexation
of chiral ligand to a strong metal-based Lewis acid.9 In general,
the donor properties of the ligand result in the attenuation of
electrophilicity of the metal center.10 Therefore, the chiral
catalyst should be preformed or else a high association constant
is necessary to avoid the achiral background reaction from the
uncomplexed Lewis acid.11
SCHEME 2
1. Auxiliary-Based Stereoselective Glycolate Aldol Reac-
tion. Chiral auxiliary-based reactions with preformed enolates
are most commonly used for the stereoselective glycolate aldol
reaction.5,6 In most cases, syn-stereoselectivity is observed using
boron enolates of chiral oxazolidinone derivatives of glycolate
esters. However, systematic studies that examine a range of
aldehydes for the chiral auxiliary-based method are rare.6
Recently, Andrus and co-workers reported a number of syn-
and anti-selective auxiliary-based glycolate aldol reactions that
took advantage of highly selective enolizations of acyclic and
cyclic glycolate esters and the high specificity of boron-mediated
aldol reactions.7 Although both diastereomers could be prepared
However, donor-acceptor interactions do not always reduce
electrophilicity of the metal center. Although electron density
is transferred from the donor to the acceptor in the coordination
interaction, it is not equally distributed since the electron density
is polarized in the complexed species toward the peripheral
ligands of the Lewis acid acceptor. Therefore, the coordination
of a polyatomic donor to a polyatomic acceptor causes an
increase of electron density on the central atom of the donor
and a decrease of electron density on the central atom of the
acceptor (Scheme 3).12 This polarization can result in the ion-
ization of one of the ligands from the acceptor atom and the
formation of a strongly Lewis acidic cation.13 Thus, it becomes
possible to employ stoichiometric amounts of a weak, achiral
Lewis acid and substoichiometric amounts of a chiral Lewis
base for asymmetric catalysis. Since the most reactive catalytic
species will be chiral, the chiral catalyst can be generated in
(3) (a) Gennari, C. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: New York, 1991; Vol. 2, Chapter 2.4. (b)
Cameron, J. C.; Paterson, I. Org. React. 1997, 51, 1. (c) Carreira, E. M. In
ComprehensiVe Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A., Yamamoto,
H., Eds.; Springer: Heidelberg, 1999; Vol. 3, Chapter 29.1. (d) Modern Aldol
Reactions; Mahrwald, R., Ed.; Wiley-VCH: Weinheim, Germany, 2004. (e)
Nelson, S. G. Tetrahedron: Asymmetry 1998, 9, 357–389. (f) Machajewski, T. D.;
Wong, C.-H. Angew. Chem., Int. Ed. 2000, 39, 1352–1374.
(4) (a) Kobayashi, S.; Furuya, M.; Ohtsubo, A.; Mukaiyama, T. Tetrahedron:
Asymmetry 1991, 2, 635–638. (b) Kobayashi, S.; Fujishita, Y.; Mukaiyama, T.
Chem. Lett. 1990, 19, 1455–1458.
(5) (a) Evans, D. A.; Bender, S. L.; Morris, J. J. Am. Chem. Soc. 1988, 110,
2506–2526. (b) Jones, T. K.; Mills, S. G.; Reamer, R. A.; Askin, D.; Desmond,
R.; Volante, R. P.; Shinkai, I. J. Am. Chem. Soc. 1989, 111, 1157–1159. (c) Ku,
T. W.; Kondrad, K. H.; Gleason, J. G. J. Org. Chem. 1989, 54, 3487–3491. (d)
Evans, D. A.; Kaldor, S. W.; Jones, T. K.; Clardy, J.; Stout, T. J. J. Am. Chem.
Soc. 1990, 112, 7001–7031. (e) Evans, D. A.; Gage, J. R.; Leighton, J. L.; Kim,
A. S J. Org. Chem. 1992, 57, 1961–1963. (f) Saika, H.; Fruh, T.; Iwasaki, G.;
Koizumi, S.; Mori, I.; Hayakawa, K. Bioorg. Med. Chem. Lett. 1993, 3, 2129–
2134. (g) Andrus, M. B.; Schreiber, S. L. J. Am. Chem. Soc. 1993, 115, 10420–
10421. (h) Evans, D. A.; Barrow, J. C.; Leighton, J. L.; Robichaud, A. J.; Sefkow,
M. J. Am. Chem. Soc. 1994, 116, 12111–12112. (i) Rudge, A. J.; Collins, I.;
Holmes, A. B.; Baker, R. Angew. Chem., Int. Ed. Engl. 1994, 33, 2320–2322.
(j) Martin, S. F.; Dodge, J. A.; Burgess, L. E.; Limberakis, C.; Hartmann, M.
Tetrahedron 1996, 52, 3229–3246. (k) Crimmins, M. T.; Choy, A. L. J. Org.
Chem. 1997, 62, 7548–7549. (l) Nakamura, Y.; Hirata, M.; Kuwano, E.;
Taniguchi, E. Biosci. Biotechnol. Biochem. 1998, 62, 1550–1554. (m) Crimmins,
M. T.; Choy, A. L. J. Am. Chem. Soc. 1999, 121, 5653–5660. (n) Haight, D.;
Birrell, H. C.; Cantello, B. C. C.; Eggleston, D. S.; Haltiwanger, R. C.; Hindley,
R. M.; Ramaswany, A.; Stevens, N. C. Tetrahedron: Asymmetry 1999, 10, 1353–
1367. (o) Crimmins, M. T.; Tabet, E. A. J. Am. Chem. Soc. 2000, 122, 5473–
5476. (p) Crimmins, M. T.; King, B. W.; Zuercher, W. J.; Choy, A. L. J. Org.
Chem. 2000, 65, 8499–8509. (q) Zhang, W.; Carter, R. G.; Yokochi, A. F. T. J.
Org. Chem. 2004, 69, 2569–2572.
(8) (a) Kobayashi, S.; Kawasuji, T. Synlett 1993, 12, 911–913. (b) Kobayashi,
S.; Kawasuji, T. Tetrahedron Lett. 1994, 35, 3329–3332. (c) Kobayashi, S.;
Horibe, M J. Am. Chem. Soc. 1994, 116, 9805–9806. (d) Kobayashi, S.; Hayashi,
T. J. Org. Chem. 1995, 60, 1098–1099. (e) Kobayashi, S.; Horibe, M.;
Matsumura, M. Synlett 1995, 6, 675–676. (f) Kobayashi, S.; Horibe, M.
Tetrahedron: Asymmetry 1995, 6, 2565–2569. (g) Kobayashi, S.; Horibe, M.;
Hachiya, I. Tetrahedron Lett. 1995, 36, 3173–3176. (h) Kobayashi, S.; Horibe,
M. Chem. Eur. J. 1997, 3, 1472–1481.
(9) Yamamoto, H. Lewis Acids in Organic Synthesis; Wiley-VCH: Weinheim,
Germany, 2001.
(10) (a) Suzuki, I.; Yamamoto, Y. J. Org. Chem. 1993, 58, 4783–4784. (b)
Santelli, M.; Pos, J.-M. Lewis Acids and SelectiVity in Organic Synthesis; CRC
Press: Boca Raton, FL, 1996.
(6) (a) Li, Z.; Wu, R.; Michalczyk, R.; Dunlap, R. B.; Odom, J. D.; Silks,
L. A., III J. Am. Chem. Soc. 2000, 122, 386–387. (b) Crimmins, M. T.;
McDougall, P. J. Org. Lett. 2003, 5, 591–594. (c) Hoover, T. R.; Hitchcock,
S. R. Tetrahedron: Asymmetry 2003, 14, 3233–3241.
(11) Gauthier, D. R., Jr.; Carreira, E. M Angew. Chem., Int. Ed. Engl. 1996,
35, 2363–2365.
(12) (a) Gutman, V. The Donor-Acceptor Approach to Molecular Interactions;
Plenum Press: New York, 1978. (b) Jensen W. B. The Lewis Acid-Base Concepts;
Wiley-Interscience: New York, 1980; Chapter 4.
(7) (a) Andrus, M. B.; Sekhar, B. B. V. S.; Turner, T. M.; Meredith, E. L.
Tetrahedron Lett. 2001, 42, 7197–7201. (b) Andrus, M. B.; Sekhar, B. B. V. S.;
Meredith, E. L.; Dalley, N. K. Org. Lett. 2000, 2, 3035–3037. (c) Andrus, M. B.;
Mendenhall, K. G.; Meredith, E. L.; Sekhar, B. B. V. S. Tetrahedron Lett. 2002,
43, 1789–1792.
(13) Bassindale, A. R.; Glynn, S. J.; Tayor, P. G. In The Chemistry of
Organosilicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; Wiley: Chichester,
1998; Vol. 2; pp 495-511.
J. Org. Chem. Vol. 73, No. 12, 2008 4583

Denmark and Chung
situ without concern for an achiral background reaction pro-
moted by the achiral Lewis acid.
methylation of carboxylic acids with diazomethane in high
yields. tert-Butyldimethylsilyl-protected methyl glycolate 3e was
prepared by silylation of commercially available methyl gly-
colate.
SCHEME 3
SCHEME 4
Other glycolate esters were synthesized by DCC-DMAP
coupling of commercially available R-methoxyacetic acid or
R-benzyloxyacetic acid with the corresponding alcohols (Scheme
5). The reactions generally proceeded smoothly and in high
yields. The coupling with 3-ethyl-3-pentanol was the only
sluggish reaction and provided the ester 3i in moderate yield.
Studies from these laboratories have demonstrated the utility
of this conceptually novel, Lewis base catalyzed, Lewis acid
mediated process in the addition of various nucleophiles
including silyl enol ethers, silyl dienol ethers, silyl ketene acetals,
N,O-silyl ketene acetals, and isocyanides to a wide range of
aldehydes utilizing the weakly Lewis acidic SiCl4 and Lewis
basic bisphosphoramide catalyst (R,R)-1.14 The primary objec-
tive of this study is to extend the scope of nucleophile to the
glycolate-derived silyl ketene acetals and thereby develop a
catalytic, stereoselective glycolate aldol reaction. From the
previous studies on other aldol processes it was known that all
of the substituents on the double bond of R-substituted silyl
keteneacetalsplayedimportantrolesincontrollingstereoselectivity.14d,iThus,
one of the major challenges was to elucidate the influence of
various structural features of the R-alkoxy silyl ketene acetal
including the protecting group of the R-hydroxyl function, the
alkyl group of the ester, and the silyl group of the ketene acetal
on the stereoselectivity.15
SCHEME 5
Results
1. Preparation of Glycolate-Derived Silyl Ketene Acetals.
A number of silyl ketene acetals were prepared from various
alkyl R-alkoxyacetates with either trimethylsilyl or tert-butyl-
dimethylsilyl groups. Methyl glycolates with a range of R-alkoxy
groups were prepared from chloroacetic acid (Scheme 4). The
ethers were formed by the reaction of chloroacetic acid with
various sodium alkoxides in refluxing THF to afford the
R-protected glycolic acids 2b-d in good yields. Then, the
corresponding methyl esters 3b-d were obtained by the
Although the enolization of glycolates is typically effected
with lithium amide bases such as LDA or LHMDS,16 it was
found that commercially available KHMDS serves as an
excellent base for the enolization (Scheme 6). Thus, various
glycolates were enolized with KHMDS and the enolates were
trapped with trimethylsilyl chloride to afford trimethylsilyl
ketene acetals 4b-f in high yields. The products could be
purified by distillation and stored for a few months at -15 °C
without significant decomposition. The methyl R-methoxyac-
etate-derived trimethylsilyl ketene acetal 4a had to be sacrifi-
cially distilled to remove HMDS byproduct because the boiling
points of 4a and HMDS were very similar. All of the
trimethylsilyl ketene acetals were obtained with very high
(14) (a) Denmark, S. E.; Barsanti, P. A.; Wong, K.-T.; Stavenger, R. A. J.
Org. Chem. 1998, 63, 2428–2429. (b) Denmark, S. E.; Wynn, T. J. Am. Chem.
Soc. 2001, 123, 6199–6200. (c) Denmark, S. E.; Fan, Y. J. Am. Chem. Soc.
2002, 124, 4233–4235. (d) Denmark, S. E.; Wynn, T.; Beutner, G. L. J. Am.
Chem. Soc. 2002, 124, 13405–13407. (e) Denmark, S. E.; Fan, Y. J. Am. Chem.
Soc. 2003, 125, 7825–7827. (f) Denmark, S. E.; Beutner, G. L. J. Am. Chem.
Soc. 2003, 125, 7800–7801. (g) Denmark, S. E.; Heemstra, J. R., Jr Org. Lett.
2003, 5, 2303–2306. (h) Denmark, S. E.; Heemstra, J. R., Jr. Synlett 2004, 13,
2411–2416. (i) Denmark, S. E.; Beutner, G. L.; Wynn, T.; Eastgate, M. D. J. Am.
Chem. Soc. 2005, 127, 3774–3789. (j) Denmark, S. E.; Fan, Y. J. Org. Chem.
2005, 70, 9667–9676. (k) Denmark, S. E.; Bui, T. J. Org. Chem. 2005, 70, 10190–
10193. (l) Denmark, S. E.; Heemstra, J. R., Jr J. Am. Chem. Soc. 2006, 128,
1038–1039. (m) Denmark, S. E.; Heemstra, J. R., Jr J. Org. Chem. 2007, 72,
56685688. (n) Denmark, S. E.; Fujimori S. In Modern Aldol Reactions;
Mahrwald, R., Ed.; Wiley-VCH: Weinheim, Germany, 2004; Chapter 7.
(15) For a preliminary communication, see: Denmark, S. E.; Chung, W.-j.
Angew. Chem., Int. Ed. 2008, 47, 1890–1892.
(16) (a) Barrish, J. C.; Lee, H. L.; Baggiolini, E. G.; Uskokvic; M, R. J.
Org. Chem. 1987, 52, 1372–1375. (b) Gould, T. J.; Balestra, M.; Wittmann,
M. D.; Gary, J. A.; Rossano, L. T.; Kallmerten, J. J. Org. Chem. 1987, 52,
3889–3901. (c) Sato, T.; Tsunekawa, H.; Kohama, H.; Fujisawa, T. Chem. Lett.
1986, 15, 1553–1556. (d) Fujisawa, T.; Tajima, K.; Sato, T. Chem. Lett. 1984,
13, 1669–1672. (e) Burke, S. D.; Fobare, W. F.; Pacofsky, G. J. J. Org. Chem.
1983, 48, 5221–5228. (f) Kallmerten, J.; Gould, T. J. Tetrahedron Lett. 1983,
24, 5177–5180. (g) Annunziata, R.; Cinquini, M.; Cozzi, F.; Borgia, A. L. J.
Org. Chem. 1992, 57, 6339–6342. (h) Mukaiyama, T.; Uchiro, H.; Shiina, I.;
Kobayashi, S. Chem. Lett. 1990, 19, 1019–1022.
4584 J. Org. Chem. Vol. 73, No. 12, 2008

Lewis Base ActiVation of Lewis Acids
TABLE 1. Glycolate Aldol Reactions with Silyl Ketene Acetals Derived from r-Alkoxy Methyl Acetatesa
entry
R1
time,b min
product
yield,c %
syn/antid
er (syn)e
er (anti)e
1
2
3
4
5
Me (4a)
i-Pr (4b)
t-Bu (4c)
PhMe2C (4d)
TBS (4e)
0.5
0.5
1.5
10.0
0.5
6aa
6ba
6ca
6da
6ea
98
95
93
98
95
57:43
86:14
99:1
99:1
98:2
73.6:26.4
79.8:20.2
93.4:6.6
96.4:3.6f
95.3:4.7f
81.5:18.5
77.0:23.0
ND
ND
ND
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of SiCl4, and 0.1 equiv of i-Pr2NEt at 0.2 M concentration. b Monitored in situ by
ReactIR. c Yields of chromatographically homogeneous material. d Determined by 1H NMR analysis of HC(2) or HC(3). e Determined by chiral
stationary phase, supercritical fluid chromatography (CSP-SFC). f (2R,3S) absolute configuration.20
SCHEME 7
geometrical purities. The major geometrical isomers were
assigned to the Z configuration by NOESY1D experiments.
SCHEME 6
tert-Butyldimethylsilyl ketene acetals 4g-m were also ob-
tained in very good yields with high Z-selectivities by a similar
protocol, although the geometrical purities were occasionally
diminished (Scheme 7). In the case of high molecular weight
ketene acetals such as 4m-o, prolonged distillation at high
temperature resulted in significant reduction of yield and purity
most likely because of a retro-ene side reaction.17 Thus, the
by in situ IR spectroscopic analysis. The reactivity of 4a was
ketene acetal 4m had to be purified by bulb-to-bulb distillation
comparable to the reactivity of simple acetate- or propanoate-
within a short period of time. Unfortunately, it was not possible
derived silyl ketene acetals.14d,i The aldol addition of 4a to 5a
to obtain 4n and 4o with high purity. With the current protocol,
in the presence of 1 mol % of the chiral bisphosphoramide
a clear advantage of KHMDS over lithium amide bases was
catalyst (R,R)-1 at -70 °C was complete in 30 s (entry 1).18
observed. For the tert-butyldimethylsilyl protection of lithiated
esters, it is necessary to use additives such as HMPA or DMPU
for rapid silylation. However, tert-butyldimethylsilyl ketene
acetals could be synthesized without such additives by employ-
ing KHMDS presumably because the potassium salt is more
reactive.
2. Lewis Base Catalyzed Glycolate Aldol Reactions.
However, whereas very high anti-diastereoselectivity and enan-
tioselectivity were observed from the E- or Z-propanoate-derived
silyl ketene acetal,14d,i 4a afforded a nearly 1:1 mixture of
diastereomers of 6aa with only moderate enantioselectivity for
each diastereomer. Interestingly, the stereoselectivities could be
improved dramatically by increasing the size of the R-alkoxy
substituent. The methyl R-isopropoxyacetate-derived trimeth-
2.1. Aromatic Aldehydes. 2.1.1. Effect of the Size of r-Alkoxy
ylsilyl ketene acetal 4b, which was as reactive as 4a, provided
Substituents on the Stereoselectivities. The silyl ketene acetals
moderate but improved diastereoselectivity (entry 2). Surpris-
derived from methyl R-alkoxyacetates were employed in the
aldol reactions with benzaldehyde (5a) to investigate the effect
ingly, the configuration of major diastereomer of 6ba was
assigned as syn (see section 2.1.3), which is opposite to the
of the size of R-alkoxy substituents on the stereoselectivities
previously observed anti-selectivity from propanoate aldol
(Table 1). Initially, the reaction with methyl R-methoxyacetate-
reaction under similar reaction conditions. Furthermore, the aldol
derived trimethylsilyl ketene acetal 4a was conducted under the
addition of methyl R-tert-butoxyacetate-derived trimethylsilyl
ketene acetal 4c resulted in very high syn-selectivity as well as
improved enantioselectivity for the syn-diastereomer of 6ca with
conditions previously developed for the aldol reactions with
simple silyl ketene actals.14d,i Reaction progress was monitored
(17) tert-Butyldimethylsilyl R-benzyloxyacetate was formed as a side product
after distillation.
(18) For a discussion of the role of Hu¨nig base, see ref 14g.
J. Org. Chem. Vol. 73, No. 12, 2008 4585

Denmark and Chung
TABLE 2. Syn-Selective Glycolate Aldol Reactions with Aromatic
only slightly reduced reaction rate (entry 3). The enantioselec-
tivity could be optimized further by introduction of even bulkier
groups such as the cumyl or tert-butyldimethylsilyl group on
the R-oxygen substituent. The reactivity of the methyl R-cumy-
loxyacetate-derived trimethylsilyl ketene acetal 4d was sub-
stantially lower than other silyl ketene acetals. Nevertheless,
the aldol addition of 4d was complete in only 10 min. Moreover,
the best diastereoselectivity and enantioselectivity within the
series could be achieved (entry 4). The methyl R-tert-butyldim-
ethylsiloxyacetate-derived trimethylsilyl ketene acetal 4e also
provided high diastereoselectivity and enantioselectivity without
attenuation of reactivity (entry 5).19 The reactivity of 4e was
comparable to that of 4a and 4b. The absolute configuration of
the aldol product 6ea was assigned to be (2R,3S) by comparison
to the optical rotation of 6ea to a literature value.20
Aldehydesa
entry
R4
time, h product yield,b
%
drc
erd
1
2
3
4
5
6
7
8
9
Ph (5a)
0.5
0.5
1.5
0.5
0.5
9.0
0.5
1.5
0.5
6da
6db
6dc
6dd
6de
6df
6dg
6dh
6di
87
97
98
93
99
93
94
93
95
99:1 96.6:3.4e
>99:1 97.5:2.5
99:1 97.7:2.3
>99:1 97.2:2.8
99:1 98.0:2.0
98:2 96.8:3.2
>99:1 98.5:1.5
99:1 96.1:3.9
98:2 72.2:27.8
4-Me-C6H4 (5b)
4-MeO-C6H4 (5c)
4-Cl-C6H4 (5d)
4-CF3-C6H4 (5e)
2-Me-C6H4 (5f)
2-naphthyl (5g)
1-naphthyl (5h)
2-benzofuryl (5i)
A number of aromatic aldehydes with diverse structural and
electronic properties were surveyed with the most selective silyl
ketene acetal 4d (Table 2). The electronic nature of the aldehyde
had little effect on reactivity and selectivity. Electron-rich
aldehydes such as 4-tolualdehyde (5b) and 4-methoxybenzal-
dehyde (5c) provided high chemical yields as well as high
diastereoselectivities and enantioselectivities although the re-
activity of 5c was slightly lower (entries 2 and 3). Similarly,
electron-poor aldehydes such as 4-chlorobenzaldehyde (5d) and
4-trifluoromethylbenzaldehyde (5e) also gave high stereoselec-
tivities and high chemical yields within 0.5 h (entries 4 and 5).
However, the steric encumbrance in the aromatic aldehyde had
a significant impact on the reactivity. The aldol reaction with
the sterically hindered 2-tolualdehyde (5f) was very slow
compared to the other aromatic aldehydes (entry 6). The reaction
required 9 h for completion. Gratifyingly, the stereoselectivities
remained high in spite of the steric hindrance. 2-Naphthaldehyde
(5g) and 1-naphthaldehyde (5 h) also behaved well and afforded
high yields and stereoselectivities (entries 7 and 8). Unfortu-
nately, 2-benzofuraldehyde (5i) was not a suitable substrate for
this reaction. Despite the high yield and diastereoselectivity,
the enantioselectivity was only moderate (entry 9).
2.1.2. Effect of the Size of the Ester on Stereoselectivity.
To evaluate the effect of the size of the ester residue, the silyl
ketene acetals derived from alkyl R-methoxyacetates were also
employed in the aldol reactions with benzaldehyde (5a) (Table
3). Again, a remarkable influence on the diastereoselectivity was
observed by increasing the size of the ester. Simply by switching
from the methyl ester-derived silyl ketene acetals to tert-butyl
ester-derived silyl ketene acetal 4f, the stereochemical course
reVersed to give high anti-diastereoselectivity (entry 1)! Interest-
ingly, the size of the silyl group also had a noticeable effect on
the stereoselectivities. By employing a tert-butyldimethylsilyl
group instead of trimethylsilyl group, the anti-diastereoselec-
tivity and the enantioselectivity of anti-diastereomer were
enhanced (entry 2). A further improvement of enantioselectivity
was achieved by the use of the 3-methyl-3-pentyl R-methoxy-
acetate-derived tert-butyldimethylsilyl ketene acetal 4h (entry
3). The enantioselectivity turned out to be quite sensitive to
small changes in the ester group. If the group was too bulky
such as 3-ethyl-3-pentyl ester, the enantioselectivity decreased
slightly (entry 4). Also, the 3-pentyl R-methoxyacetate-derived
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of
SiCl4, and 0.1 equiv of i-Pr2NEt at 0.1 M concentration. The silyl
ketene acetal was added as a 0.24 M solution over 15 min. b Yields of
analytically pure material. c Determined by 1H NMR analysis of HC(2)
or HC(3). d Determined by CSP-SFC. e (2R,3S) absolute configuration.
tert-butyldimethylsilyl ketene acetal 4j, which has one fewer
methyl groups than 4h, afforded significantly diminished
diastereoselectivity and enantioselectivity (entry 5). Finally, the
neopentyl R-methoxyacetate-derived tert-butyldimethylsilyl ketene
acetal 4k was tested. Although 4k showed much higher
reactivity compared to other bulky ester-derived silyl ketene
acetals, the resulting diastereoselectivity and enantioselectivity
were moderate, presumably because of the attenuation of steric
bulk which seemed necessary to achieve the high stereoselec-
tivities (entry 6).
Although satisfactory yields and stereoselectivities were
achieved for anti-diastereoselective glycolate aldol reactions with
4h, the methyl group on the R-oxygen is not a desirable group.
Therefore, 4h was modified with more readily removable groups
such as benzyl (4n) or trimethylsilylethyl (4o) groups. However,
the aldol reactions with the modified silyl ketene acetals 4n and
4o resulted in significant reduction of enantioselectivity (Scheme
8).
SCHEME 8
Having identified an optimal silyl ketene acetal for the anti-
selective glycolate aldol reaction (4h), a series of structurally
and electronically diverse aromatic aldehydes were tested. The
reactivity and selectivity profiles were very similar to the case
of syn-selective glycolate aldol reactions. From the reactions
with both electron-rich substrates, such as 4-tolualdehyde (5b)
(19) The reaction is highly exothermic. Later, it was found that the
stereoselectivities could be improved by slow addition of silyl ketene acetal as
a solution, thereby avoiding the increase of internal temperature. When the silyl
ketene acetal 4e was added as 0.24 M solution for 15 min, the aldol product 6ea
was obtained in 97% yield, 99/1 dr (syn/anti), and 96.5/3.5 er (syn).
(20) Carda, M.; Murga, J.; Falomir, E.; Gonza´lez, F.; Marco, J. A.
Tetrahedron 2000, 56, 677–683.
4586 J. Org. Chem. Vol. 73, No. 12, 2008

Lewis Base ActiVation of Lewis Acids
TABLE 3. Glycolate Aldol Reactions with Silyl Ketene Acetals Derived from Bulky Alkyl Estersa
entry
R2
time,b min
product
yield,c %
syn:antid
er (syn)e
er (anti)e
1
2
3
4
5
6
t-Bu (4f)
t-Bu (4g)
Et2MeC (4h)
Et3C(4i)
Et2CH (4j)
neo-pentyl (4k)
3.5
6.0
6.0
6.0
0.5
0.5
6fa
6fa
6ha
6ia
6ja
6ka
93
92
92f
93
97
98
4:96
1:99
1:99
2:98
8:92
17:83
90.4:9.6
ND
ND
ND
37.8:62.2
15.3:84.7
81.3:18.7
92.1:7.9
94.0:6.0
92.8:7.2
82.9:17.1
79.7:20.3
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of SiCl4, and 0.1 equiv of i-Pr2NEt at 0.1 M concentration. b Monitored in situ by
ReactIR. c Yields of chromatographically homogeneous material. d Determined by 1H NMR analysis of HC(2) or HC(3). e Determined by CSP-SFC.
f (2S,3S) absolute configuration.
TABLE 4. Anti-Selective Glycolate Aldol Reactions with Aromatic
major diastereomer of 8fa was 9.3 Hz. Subsequently, the major
diastereomer of 8fa was assigned to be the anti-configuration.
The aldol products 6ia, 6ja, and 6ka were also reduced to the
corresponding diols. The major diastereomers of those diols were
identical to the major diastereomer of 8fa by 1H NMR analysis.
Therefore, the configurations of the major diastereomers of 6ia,
6ja, and 6ka were also confirmed to be anti-configuration.
The determination of absolute configuration required a few
synthetic manipulations of representative glycolate aldol prod-
ucts (Scheme 10). The cumyl group of syn-glycolate aldol
product 6da was easily removed with in situ generated
anhydrous HCl to give the known diol 9.21 Comparison of its
optical rotation with the data reported for this compound
revealed that the absolute configuration of 9 is (2R,3S).
However, the selective removal of the methyl group of 6ha
without altering the sensitive benzylic hydroxyl group was
challenging. After an extensive survey of demethylation condi-
tions, the methyl group of 6da was successfully removed by
AlCl3 in n-BuSH.22 Although the tertiary alkyl group of the
ester was also removed under the reaction conditions, the ben-
zylic hydroxyl group remained intact. Because of the difficulty
in the purification of the dihydroxy carboxylic acid, the crude
material was directly treated with CH2N2. The resulting dihy-
droxy ester 10 was easily purified, and its absolute configuration
was subsequently assigned as (2S,3S) by comparison of its
optical rotation with the reported data.23 The 3S configuration
of both 9 and 10 imply a Re face attack of the nucleophile on
the aldehyde. This sense of asymmetric induction is consistent
with previously reported results from the related studies with
the same catalytic system.14
Aldehydesa
entry
R4
time, h product yield,b
%
drc
erd
1
2
3
4
5
6
7
8
9
Ph (5a)
0.5
0.5
3.0
0.5
0.5
18.0
0.5
5.0
0.5
6ha
6hb
6hc
6hd
6he
6hf
6hg
6hh
6hi
91e
92
93
92
96
91
93
94
97
>99:1 95.1:4.9
99:1 96.1:3.9
>99:1 98.3:1.7
99:1 95.1:4.9
>99:1 98.3:1.7
98:2 93.1:6.9
>99:1 97.4:2.6
>99:1 78.2:21.8
99:1 94.9:5.1
4-Me-C6H4 (5b)
4-MeO-C6H4 (5c)
4-Cl-C6H4 (5d)
4-CF3-C6H4 (5e)
2-Me-C6H4 (5f)
2-naphthyl (5g)
1-naphthyl (5h)
2-benzofuryl (5i)
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of
SiCl4, and 0.1 equiv of i-Pr2NEt at 0.1 M concentration. The silyl
ketene acetal was added as a 0.24 M solution over 15 min. b Yields of
analytically pure material. c Determined by 1H NMR analysis of HC(2)
or HC(3). d Determined by CSP-SFC. e (2S,3S) Absolute configuration.
and 4-methoxybenzaldehyde (5c), and electron-poor substrates,
such as 4-chlorobenzaldehyde (5d) and 4-trifluoromethylben-
zaldehyde (5e), consistently high yields, diastereoselectivities,
and enantioselectivities were obtained (Table 4, entries 2-5).
Again, a significantly decreased reaction rate was observed from
2-tolualdehyde (5f) while the stereoselectivities remained
satisfying (entry 6). Whereas 2-naphthaldehyde (5g) provided
high yield and stereoselectivities (entry 7), 1-naphthaldehyde
(5h) exhibited relatively poor reactivity and resulted in low
enantioselectivity (entry 8). In contrast to the syn-selective
glycolate aldol reaction, 2-benzofuraldehyde (5i) served as a
good substrate in the anti-selective manifold giving high yield
as well as high stereoselectivities (entry 9).
2.1.3. Assignment of Relative and Absolute Configura-
tions of Glycolate Aldol Products of Benzaldehyde. For
unambiguous determination of the relative configurations, the
glycolate aldol products from the optimization studies were
reduced to the corresponding diols 7 with lithium aluminum
hydride and then were converted to 6-membered cyclic acetals
8 (Scheme 9). The vicinal coupling constants between HC(1)
and HC(2) of the major diastereomers of 8ba and 8ca were 2.2
and 2.4 Hz, respectively. Thus, the major aldol products of 6ba
and 6ca were assigned to be the syn-configuration. On the other
hand, the coupling constant between HC(1) and HC(2) of the
2.2. Aliphatic Aldehydes. 2.2.1. Syn-Selective Glycolate
Aldol Reactions with Aliphatic Aldehydes. The attenuated
reactivity of aliphatic aldehydes has been consistently observed
in the previous studies of Lewis base catalyzed, SiCl4-promoted
carbonyl addition reactions.14 It was demonstrated by H and
1
13C NMR analysis that cyclohexanecarboxyaldehyde (5o) is
instantaneously converted to R-chloro trichlorosilyl ether 11 in
the presence of HMPA and SiCl4 (Scheme 11).14i The R-chloro
trichlorosilyl ether is unreactive toward nucleophiles and thus
(21) (a) Choudary, B. M.; Chowdari, N. S.; Madhi, S.; Kantam, M. L. Angew.
Chem., Int. Ed. 2001, 40, 4620–4623. (b) Wang, Z.-H.; Kolb, H. C.; Sharpless,
K. B. J. Org. Chem. 1994, 59, 5104–5105.
(22) Node, M.; Nishide, K.; Sai, M.; Ichikawa, K.; Fuji, K.; Fujita, E. Chem.
Lett. 1979, 8, 97–98.
(23) (a) Matthews, B. R.; Jackson, W. R.; Jacobs, H. A.; Watson, K. G.
Aust. J. Chem. 1990, 43, 1195–1214. (b) Fujita, M.; Laine, D.; Ley, S. V.
J. Chem. Soc., Perkin Trans. 1 1999, 12, 1647–1656.
J. Org. Chem. Vol. 73, No. 12, 2008 4587

Denmark and Chung
TABLE 5. Optimization of syn-Selective Glycolate Aldol Reactions
SCHEME 9
with Aliphatic Aldehydesa
entry
R1
product
yield,b
%
drc
erd
1
2
PhMe2C (4d)
TBS (4e)
6dj
6ej
trace
49
ND
90:10
ND
12.7:87.3e
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of
SiCl4, and 0.1 equiv of i-Pr2NEt at 0.4 M concentration. b Yields of
chromatographically homogeneous material. c Determined by 1H NMR
analysis of HC(2) or HC(3). d Determined by CSP-SFC. e (2R,3S)
absolute configuration.
SCHEME 11
To extend the substrate scope to aliphatic aldehydes, the aldol
reaction between the trimethylsilyl ketene acetal 4d and
hydrocinnamaldehyde (5j) was attempted (Table 5, entry 1).
Because of the low reactivity of aliphatic aldehydes, more
forcing reaction conditions such as elevated temperature, higher
concentration, and higher catalyst loading were employed for
this class of aldehydes. Disappointingly, only traces of aldol
product were observed after 20 h in the presence of 5 mol % of
(R,R)-1 even at -50 °C and 0.4 M concentration. Although the
silyl ketene acetal 4d was the most selective silyl ketene acetal
for the syn-selective aldol reactions with aromatic aldehydes, it
was also the least reactive silyl ketene acetal. Thus, the reactivity
problem could be solved simply by employing a more reactive
silyl ketene acetal. During the survey of silyl ketene acetals for
the aldol reaction with benzaldehyde, the methyl R-tert-
butyldimethylsiloxyacetate-derived trimethylsilyl ketene acetal
4e was identified as a highly reactive and stereoselective silyl
ketene acetal. Therefore, the silyl ketene acetal 4e was tested
for the aldol reaction with 5j (entry 2). Consequently, the
chemical yield was dramatically improved although it is still
moderate. Moreover, reasonable enantioselectivity and diaste-
reoselectivity were observed. To improve the yield and selectiv-
ity further, the silyl group at R-hydroxy function of 4e was
modified to other silyl groups including triethylsilyl, triisopro-
pylsilyl, tert-hexyldimethylsilyl, and dimethylphenylsilyl groups.
In addition, the corresponding triethylsilyl and tert-butyldim-
ethylsilyl ketene acetal analogues of 4e were tested. However,
these attempts were not successful. None of the modified silyl
ketene acetals gave better results than the parent methyl R-tert-
butyldimethylsiloxyacetate-derived trimethylsilyl ketene acetal
4e.
SCHEME 10
2.2.2. Anti-Selective Glycolate Aldol Reactions with Ali-
phatic Aldehydes. An approach similar to that described above
was employed for the anti-selective glycolate aldol reaction with
5j. Poor reactivity was encountered again as only trace amounts
of product were observed from the reaction of 3-methyl-3-pentyl
R-methoxyacetate-derived tert-butyldimethylsilyl ketene acetal
4h after 20 h in the presence of 5 mol % of (R,R)-1 at -50 °C
responsible for the low reactivity of aliphatic aldehydes.
Moreover, because the equilibrium of this reaction lies on the
R-chloro trichlorosilyl ether, the concentration of aldehyde is
very low. Nevertheless, highly reactive nucleophiles still could
react with these substrates at reasonable reaction rate.
4588 J. Org. Chem. Vol. 73, No. 12, 2008

Lewis Base ActiVation of Lewis Acids
TABLE 7. Anti-Selective Glycolate Aldol Reactions with Aliphatic
and 0.4 M concentration (Table 6, entry 1). Thus, more reactive
silyl ketene acetals were sought. During the survey of silyl
ketene acetals for the aldol reaction with benzaldehyde, neo-
pentyl R-methoxyacetate-derived tert-butyldimethylsilyl ketene
acetal 4k and 3-pentyl R-methoxyacetate-derived tert-butyldim-
ethylsilyl ketene acetal 4j showed very high reactivity. Even
though the previously observed selectivities from these two silyl
ketene acetals were moderate, the silyl ketene acetals 4k and
4j were tested for the glycolate aldol reaction with 5j. Gratify-
ingly, the silyl ketene acetal 4k afforded the aldol product 6kj
in good yield with high diastereoselectivity and enantioselec-
tivity (entry 2). Moreover, even better results were realized from
4j (entry 3). For further improvement of the stereoselectivities,
the silyl ketene acetal 4j was modified to more bulky silyl ketene
acetal 4l because the studies for the aromatic aldehydes
suggested that increased steric bulk on the ester alkyl group
would be beneficial for the high anti-selectivity. Consequently,
the silyl ketene acetal 4l afforded the aldol product 6lj with
exclusive diastereoselectivity and excellent enantioselectivity
(entry 4).
Aldehydesa
entry
R1
R4
product yield,b
%
drc
erd
1
2
3
4
5
6
7
Bn (4m) PhCH2CH2 (5j)
Bn (4m) BnO(CH2)5 (5k)
Bn (4m) BnOCH2CH2 (5l) 6mL
Bn (4m) BnOCH2 (5m)
Bn (4m) Me2CHCH2 (5n) 6mn
Bn (4m) c-C6H11 (5o)
Me (4l) c-C6H11 (5o)
6mj
6mk
82
89
98:2 96.4:3.6e
98:2 98.5:1.5
99:1 92.5:7.5
ND ND
>99:1 95.3:4.7
98:2 68:0:32.0
>99:1 87.7:12.3
59 (80)f
6mm trace
83
6mo
6lo
9g
46g
a Reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of
SiCl4, and 0.1 equiv of i-Pr2NEt at 0.4 M concentration. b Yields of
analytically pure material. c Determined by 1H NMR analysis of HC(2)
or HC(3). d Determined by CSP-SFC. e (2S,3S) absolute configuration.
f Reaction employed 2.0 equiv of silyl ketene acetal, 1.1 equiv of SiCl4,
and 0.1 equiv of i-Pr2NEt at 0.4
chromatographically homogeneous material.
M
concentration. g Yield of
TABLE 6. Optimization of Anti-Selective Glycolate Aldol
Reactions with Aliphatic Aldehydesa
4l instead of 4m. The aldol reaction of 4l with 5o provided
moderate yield and enantioselectivity (entry 7).
2.2.3. Assignment of Relative and Absolute Configura-
tions of Glycolate Aldol Products of Hydrocinnamaldehyde.
The relative configurations of the glycolate aldol products from
the optimization study for the anti-aldol reaction with aliphatic
aldehydes was unambiguously determined as before. Thus, the
aldol product 6kj was reduced to the corresponding diols 7kj
with lithium aluminum hydride and then was transformed to
6-membered cyclic acetal 8kj (Scheme 12). The vicinal coupling
constant between HC(1) and HC(2) of the major diastereomer
of 8kj was 8.5 Hz. Thus, the major diastereomer of 8kj
possessed the anti-configuration. The aldol products 6jj and 6lj
were also reduced to the corresponding diols. The major
diastereomers of those diols were identical to the major
diastereomer of 7kj. Consequently, the anti-configurations of
6jj and 6lj were confirmed.
entry
R2
product
yield,b
%
drc
erd
ND
94.7:5.3
96.3:3.7
97.0:3.0
1
2
3
4
Et2MeC (4h)
neo-pentyl (4k)
Et2CH (4j)
6hj
6kj
6jj
6lj
trace
81
86
ND
91:9
97:3
i-Pr2CH (4l)
88
>99:1
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of
SiCl4, and 0.1 equiv of i-Pr2NEt at 0.4 M concentration. b Yields of
chromatographically homogeneous material. c Determined by 1H NMR
analysis of HC(2) or HC(3). d Determined by CSP-SFC.
In contrast to the anti-selective glycolate aldol reactions with
aromatic aldehydes, the undesired R-methoxy group could be
replaced by the more synthetically useful benzyloxy group in
reactions with aliphatic aldehydes. The resulting silyl ketene
acetal 4m reacted with 5j to afford anti-aldol product 6mj
without significant loss of chemical yield and stereoselectivities
(Table 7, entry 1). Another primary aliphatic aldehyde, 6-ben-
zyloxyhexanal (5k), which has an ether function at the remote
position was an excellent substrate. The corresponding aldol
product 6mk was obtained with very high yield and selectivities
(entry 2). However, the reactivity of the aldehyde decreased
when the ether function was closer to the carbonyl group. In
the case of 3-benzyloxypropanal (5l), an excess of the silyl
ketene acetal was necessary for high chemical yield (entry 3).
This reactivity trend was dramatically illustrated when only trace
amounts of product were observed from the reaction with
2-benzyloxyacetaldehyde (5m) (entry 4). The branched aliphatic
aldehydes were also surveyed. The result from ꢀ-branched
isovaleraldehyde (5n) was comparable to the results from
unbranched aldehydes (entry 5). On the contrary, the yield and
enantioselectivity drops significantly when R-branched cyclo-
hexanal (5o) was employed (entry 6). The yield and enantiose-
lectivity could be improved by employing the silyl ketene acetal
SCHEME 12
A few synthetic transformations were needed to determine
the absolute configurations of glycolate aldol products 6ej and
6mj (Scheme 13). The aldol product 6ej was indirectly
correlated with another aldol product 6ep which was assigned
J. Org. Chem. Vol. 73, No. 12, 2008 4589

Denmark and Chung
TABLE 8. Syn-Selective Glycolate Aldol Reactions with Alkenyl Aldehydesa
entry
R4
time, h
product
yield,b
90
%
1,2:1,4c
syn:antid
er (syn)e
97.4:2.6f
75.3:24.7i
95.3:4.7j
1
2
3
PhCHdCH (5p)
PhCHdCMe (5q)
Me2CdCH (5r)
0.5
1.0
0.5
6ep
6eq
6er
94:6
99:1
92:8
99:1
83:17
>99:1
78g (16)h
82
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of SiCl4, and 0.1 equiv of i-Pr2NEt at 0.1 M concentration. The silyl ketene acetal
was added as a 0.24 M solution in CH2Cl2 over 15 min. b Yields of analytically pure material of 1,2-adduct. c Determined by 1H NMR analysis.
d
Determined by H NMR analysis of HC(2) or HC(3). e Determined by CSP-SFC. f (2R,3S) Absolute configuration. g Yield of syn-diastereomer. h Yield
1
of anti-diastereomer. i er of anti-diastereomer was 96.4:3.6. j er was determined after 2,4-dinitrobenzoylation of the hydroxyl group.
TABLE 9. Anti-Selective Glycolate Aldol Reactions with Alkenyl Aldehydesa
entry
R4
time, h
product
yield,b
%
1,2:1,4c
drd
ere
1
2
3
PhCHdCH (5p)
PhCHdCHMe (5q)
Me2CdCH2 (5r)
0.5
1.0
1.0
6mp
6mq
6mr
90
70
79
95:5
80:20
91:9
99:1
>99:1
78:22
98.0:2.0f
89.1:10.9
96.1:3.9
a All reactions employed 1.2 equiv of silyl ketene acetal, 1.1 equiv of SiCl4, and 0.1 equiv of i-Pr2NEt at 0.1 M concentration. The silyl ketene acetal
was added as a 0.24 M solution in CH2Cl2 over 15 min. b Yields of analytically pure material of 1,2-adduct. c Determined by 1H NMR analysis.
d
Determined by H NMR analysis of HC(2) or HC(3). e Determined by CSP-SFC. f (2S,3S) Absolute configuration.
1
to be (2R,3S) (see section 2.3.3). Hydrogenation of 6ep afforded
the product whose major stereoisomer was identical to the major
stereoisomer of 6ej by 1H NMR and CSP-SFC analyses. Thus,
the absolute configuration of 6ej was also assigned to be (2R,3S).
The aldol product 6mj was transformed to the known aldehyde
13 through the sequence of tert-butyldiphenylsilyl protection
and partial reduction to aldehyde by diisobutylaluminum hy-
dride.24 Comparison of its optical rotation with the reported data
revealed that the absolute configuration of 13 is (2S,3S). Both
6ej and 6mj have a 3S configuration at the ꢀ-hydroxyl-bearing
carbon center. The Re face attack of the nucleophile on the
aldehyde was confirmed again.
acetal that was optimal for aldol reaction of aliphatic aldehydes
could be applied to the aldol reaction of alkenyl aldehydes. The
reactions were carried out in the presence of 1 mol % of (R,R)-1
at -78 °C and 0.1 M concentration due to the high reactivity
of alkenyl aldehydes. Accordingly, the methyl R-tert-butyldim-
ethylsiloxyacetate-derived trimethylsilyl ketene acetal 4e reacted
with E-cinnamaldehyde (5p) to afford syn-aldol product 6ep in
high yield with high diastereoselectivity and enantioselectivity
(Table 8, entry 1). The reaction of the silyl ketene acetal 4e
and R-methylcinnamaldehyde (5q), which is a typically prob-
lematic substrate under Lewis base catalysis, proceeded with
moderate diastereoselectivity (entry 2). Although the minor
diastereomer of 6eq was formed with high enantiopurity, the
enantioselectivity of the major diastereomer was poor. Finally,
3-methyl-2-butenal (5r) gave the aldol product 6er with
excellent diastereoselectivity and enantioselectivity (entry 3).
Along with major 1,2-adducts, small amounts of 1,4-adducts
were observed in all cases.
SCHEME 13
2.3.2. Anti-Selective Glycolate Aldol Reactions with Alk-
enyl Aldehydes. The 2,4-dimethyl-3-pentyl R-benzyloxyacetate-
derived tert-butyldimethylsilyl ketene acetal 4m was also
suitable for the aldolizations with alkenyl aldehydes (Table 9).
Under the reaction conditions, the silyl ketene acetal 4m rapidly
reacted with 5p to produce the corresponding anti-diastereomer
of aldol product 6mp with high stereoselectivity (entry 1).
Extremely high diastereoselectivity was also observed from the
reaction of the silyl ketene acetal 4m with 5q (entry 2).
However, the enantioselectivity was only moderate. On the other
hand, the 5r gave the aldol product 6mr with low diastereo-
selectivity whereas the enantioselectivity was high (entry 3).
The desired aldol products were again accompanied by small
amounts of the 1,4-adduct.
2.3. Alkenyl Aldehydes. 2.3.1. Syn-Selective Glycolate
Aldol Reactions with Alkenyl Aldehydes. The silyl ketene
The direct product of 1,2-addition is a trichlorosilyl ether,
which is hydrolyzed upon aqueous workup to afford the
(24) Evans, D. A.; Glorius, F.; Burch, J. D. Org. Lett. 2005, 7, 3331–3333.
4590 J. Org. Chem. Vol. 73, No. 12, 2008

Lewis Base ActiVation of Lewis Acids
corresponding alcohol. Thus, the direct product of 1,4-addition
is most likely a trichlorosilyl enol ether (Scheme 14). However,
whereas no trans-silylation of the trichlorosilyl group with the
trimethylsilyl or tert-butyldimethylsilyl groups of the ketene
acetal has been observed in the case of 1,2-addition, the trans-
silylated 1,4-adduct 15 was isolated as a mixture of diastereo-
mers in 14% yield from the reaction of 5q and 4m.
were also demonstrated. Reactions of the structurally and
electronically diverse aldehydes provided the glycolate aldol
products with high diastereoselectivity and enantioselectivity.
The relative and absolute configurations were unambiguously
assigned. In every single case, the Re-face attack of the
nucleophiles was confirmed.
Discussion
SCHEME 14
1. High Z-Selectivity of Formation of Glycolate-Derived
Silyl Ketene Acetals. The enolization of simple propanoate
esters can be controlled by changing base or solvent system.26
Thus, it is possible to obtain both Z- and E-propanoate-derived
silyl ketene acetals with good selectivity. On the contrary, the
enolization of glycolate esters generally affords Z-enolates with
high selectivity regardless of reaction conditions.16 Only a few
examples of E-selective enolization of glycolate esters with
limited success have been reported.27 The high Z-selectivity is
generally attributed to the coordination of the R-oxygen to the
countercation of the base which is typically lithium. Thus, the
Z-selectivity is attenuated if the R-oxygen bears a bulky
protecting group. Enolization of 3d and 3e, which bear bulky
groups on the R-oxygens, with LHMDS affords the correspond-
ing silyl ketene acetals 4d and 4e with 85:15 and 75:25 (Z/E)
selectivities, respectively (Scheme 16). However, the enolization
of 3d and 3e with KHMDS afforded the silyl ketene acetal 4d
and 4e with almost exclusive Z-selectivity (Scheme 6) in spite
of poor coordinating ability of potassium cation. These results
can be explained by thermodynamic equilibration of enolate
anion. It is known that the Z-selectivity of the enolization of
glycolate esters could be further improved under thermodynamic
conditions.16 For example, the equilibration can be initiated by
the addition of HMPA, which weakens the Li-O bond, to afford
a more stable Z-enolate with improved selectivity. Similarly,
potassium enolates can undergo equilibration even in the absence
of additives because of the weak K-O bond. Therefore, higher
Z-selectivity was obtained from the enolization with KHMDS
than the enolization with LHMDS.
2.3.3. Assignment of Relative and Absolute Configura-
tions of Glycolate Aldol Products of E-Cinnamaldehyde. To
assign the absolute configuration, 6ep was transformed to the
known triol 18 through the sequence of TBAF-promoted TBS
deprotection and lithium aluminum hydride reduction (Scheme
15). Comparison of its optical rotation with the reported data
revealed that the absolute configuration of 18 is (2R,3S).25 The
aldol product 6mp was indirectly correlated with another aldol
product 6mj which was assigned to be (2S,3S) (see section
2.2.3). The hydrogenation of 6mp afforded the product whose
major stereoisomer was identical to that of 6mj as confirmed
by 1H NMR and CSP-SCF analyses. Thus, the absolute
configuration of 6mp was also assigned to be (2R,3S). The 3S
configuration at the ꢀ-hydroxyl-bearing carbon centers of 6ep
and 6mp confirmed the Re face attack of the nucleophile.
SCHEME 16
SCHEME 15
2. 1,4-Addition to Alkenyl Aldehydes. The formation of a
1,4-adduct from the addition of glycolate-derived silyl ketene
acetals to alkenyl aldehydes (Table 8 and 9) was unexpected,
as it had previously not been detected from the addition of
acetate- or propanoate-derived silyl ketene acetal. The steric
hindrance of reacting carbon of silyl ketene acetals in this study
is probably responsible for the observation of 1,4-addition. The
glycolate-derived silyl ketene acetal has a bulkier substituent
at the reacting carbon compared to previously employed
The most unique feature of Lewis base catalyzed glycolate
aldol reactions is the accessibility of both diastereomers with
the same catalytic system. Through an extensive and systematic
examination of the structure of the silyl ketene acetal, it was
possible to optimize the structure of the silyl ketene acetal for
both syn- and anti-manifolds. The generality of these reactions
(26) Ireland, R. E.; Wipf, P.; Armstrong, J. D., III J. Org. Chem. 1991, 56,
650–657.
(27) (a) Sparks, M. A.; Panek, J. S J. Org. Chem. 1991, 56, 3431–3438. (b)
Hattori, K.; Yamamoto, H. J. Org. Chem. 1993, 58, 5301–5303. (c) Hattori, K.;
Yamamoto, H. Tetrahedron 1994, 50, 3099–3112.
(25) Evans, P.; Johnson, P.; Tayler, R. J. K. Eur. J. Org. Chem. 2006, 7,
1740–1754.
J. Org. Chem. Vol. 73, No. 12, 2008 4591

Denmark and Chung
nucleophiles such as acetate- or propanoate-derived silyl ketene
acetal. As a result, the reaction with the carbonyl carbon of
alkenyl aldehyde, which is located inside of the congested
catalyst pocket (vide infra), becomes relatively difficult. There-
fore, the minor reaction takes place at the more accessible
ꢀ-carbon of alkenyl aldehyde.
3. Rationalization of the Observed Diastereoselectivity.
To gain an understanding of the substituent dependence on
diastereoselectivity, several possible transition structures were
considered. The fact that both syn- and anti-diastereomers were
obtained from the same Z-enolate rules out the possibility of a
closed, chairlike transition structure.3 Instead the aldol additions
are likely proceeding through an open transition structure.3 Six
limiting, staggered, open transition structures for the addition
of propanoate derived silyl ketene acetal to aldehydes have been
carefully analyzed previously to explain the observed anti-
diastereoselectivity.14i Following Heathcock and co-worker’s
analysis,28 minimization of the unfavorable dipole-dipole
interaction was considered as one of the major stabilizing factors.
Consequently, the observed anti-diastereoselectivity for this
propanoate aldol reaction was explained by the two antiperipla-
nar (ap) transition structures where the position of the R-sub-
stituent is the major stereochemistry-determining factor.
A similar argument can be applied for the glycolate aldol
reaction. However, because of the additional dipole of the
R-alkoxy substituent, none of the transition structures has
substantially more favorable dipole-dipole interaction than the
others (Figure 1). Although -scsyn and +scanti transition
structures (sc ) synclinal) have most unfavorable dipole-dipole
interactions between the carbonyl group of aldehyde and the
alkoxy substituents of the silyl ketene acetal, the other four
transition structures also possess unfavorable dipole-dipole
interactions. Moreover, the silyl ketene acetals in this study have
bulkier substituents than the propanoate-derived silyl ketene
acetals in the previous studies. Consequently, steric interactions
may determine the relative stability of the transition structures
more than the dipole-dipole interactions. Therefore, it is
necessary to carefully analyze the steric interactions of the silyl
ketene acetal and the catalyst in all six limiting staggered open
transition structures.
3.1. Transition-State Model for Syn-Selective Glycolate
Aldol Reactions. First, the most stable conformation of the
complex of (R,R)-1, SiCl3+, and benzaldehyde was obtained by
examining the dihedral angle around the bond between the
central silicon and the oxygen atom of aldehyde.29 Then, the
reaction coordinates of the glycolate aldol reaction was calcu-
lated by mapping the distance between the R-carbon of silyl
ketene acetals and the carbonyl carbon of the aldehyde. The
structure at the highest position of the reaction coordinates was
refined and verified by the presence of a single negative vibration
after vibrational transition calculations. The most stable con-
formation of the silyl ketene acetal in each transition structure
was obtained by calculating the Gibbs free energy of each
transition structure with four limiting conformations of silyl
ketene acetals (Figure 2)30 at 200 K. These calculations were
performed by the PC version of CAChe 6.1 with the PM3 basis
set. To minimize the total number of possible conformations,
the structure of methyl R-tert-butoxyacetate-derived trimethyl-
FIGURE 1. Overall dipoles of six limiting, staggered, open transition
structures.
silyl ketene acetal 4c was employed instead of more complex
structure of methyl R-cumyloxyacetate-derived trimethylsilyl
ketene acetal 4d.
FIGURE 2. Four limiting conformations of silyl ketene acetals.
The computationally generated transition structures suggest
that the major steric interactions come from two naphthyl rings
of each binaphthyl group and the two chlorine atoms at the
central silicon cation (Figure 3). While the apsyn, -scsyn, and
-scanti transition structures suffer from severe steric encum-
brance between the chlorine atoms of the central silicon cation
and one of the alkoxy substituents of the silyl ketene acetal,
the +scanti and apanti transition structures are destabilized by
the steric interaction between one of the naphthyl ring of the
catalyst and one of the bulky substituents of the silyl ketene
acetal. The +scsyn transition structure possesses the least
unfavorable steric interaction between the catalyst and the silyl
ketene acetal because the bulky tert-butyl group and the
trimethylsilyl group of the silyl ketene acetal are located away
from the catalyst. Therefore, the observed syn-diastereoselec-
tivity can be explained by +scsyn transition structure.
3.2. Transition-State Model for Anti-Selective Glycolate
Aldol Reactions. A similar analysis can be made for the
transition structures of glycolate aldol reactions of bulky tertiary
ester-derived silyl ketene acetals (Figure 4). The stable confor-
mation of the silyl ketene acetal in each transition structure was
also obtained by calculating the Gibbs free energy of each
transition state with all possible limiting conformations of silyl
ketene acetals at 200 K as above. To minimize the total number
of possible conformations, the structure of tert-butyl R-meth-
oxyacetate-derived trimethylsilyl ketene acetal 4f was used
instead of the more complex structure of 3-methyl-3-pentyl
R-methoxyacetate-derived tert-butyldimethylsilyl ketene acetal
4h. Whereas the apanti transition structure has only small steric
interaction between the catalyst and the methoxy substituent of
the silyl ketene acetal, the other transition structures suffer from
(28) Heathcock, C. H.; Davidsen, S. K.; Hug, K. T.; Flippin, L. A. J. Org.
Chem. 1986, 51, 3027–3037.
(29) The calculated coordinates of the complex (R,R)-1·SiCl3 were obtained
from those previously published, see ref 14i.
(30) Wilcox, C. S.; Babston, R. E. J. Org. Chem. 1984, 49, 1451–1453.
4592 J. Org. Chem. Vol. 73, No. 12, 2008

Lewis Base ActiVation of Lewis Acids
FIGURE 3. Six open transition structures for glycolate aldol reactions with a bulky R-alkoxy group.
severe steric encumbrance between the trichlorosilyl cation-
bisphosphoramide complex and one of the alkoxy substituents
of the silyl ketene acetal. Therefore, the observed anti-
diastereoselectivity can be explained by apanti transition structure.
The apanti transition structure also explains the necessity of the
methyl group on the R-oxygen for the high diastereoselectivity.
Because the R-alkoxy group and the catalyst are closely located
in apanti transition structure, bulky R-alkoxy group would
destabilize the transition structure. Thus, the introduction of
benzyl group or trimethylsilylethyl group on the R-oxygen
resulted in low diastereoselectivity.
4. Reactivity Trends. 4.1. Syn-Selective Glycolate Aldol
Reactions. The observed reactivity trend of silyl ketene acetal
of syn-selective glycolate aldol reactions can be rationalized by
the +scsyn transition structure (Figure 3). Because the R-alkoxy
group is present in relatively vacant space of the +scsyn transition
structure, the variation of the size of the R-alkoxy group would
not have a significant influence on the reaction rate until the
size of the R-alkoxy group becomes large enough to experience
steric repulsion. Thus, the observed reaction rates of the silyl
ketene acetals 4a-c and 4e are similar whereas the bulkiest silyl
ketene acetal 4d showed an attenuated reaction rate.
rate would be inversely proportional to the size of the ester.
Consequently, the primary or secondary ester-derived silyl
ketene acetals such as 4j and 4k exhibited greater reactivity
than those that were derived from tertiary esters. The ability of
the anti-selective reagents to combine with aliphatic aldehydes
most likely results from the reduced steric demand of the alkoxy
substituent on the R-carbon.
5. Reactivity Trend of Aliphatic Aldehydes with Ether
Functions. The reactivity of the aliphatic aldehyde dramatically
decreased when the ether function is closer to the carbonyl group
(Table 7, entries 2-4). This trend can be rationalized by the
equilibrium between ionized aldehyde-SiCl4 complex and
R-chloro trichlorosilyl enol ether (Scheme 11). Because the
electron-withdrawing ether function will destabilize the cationic
complex, the equilibrium position will shift to the neutral
R-chloro trichlorosilyl enol ether. Therefore, the concentration
of active aldehyde decreases further when the ether function is
located closer to the carbonyl group.
Conclusion
A general and selective Lewis base catalyzed addition of
glycolate-derived silyl ketene acetals to a wide range of
aldehydes has been developed. Both syn- and anti-diastereomers
were accessible without changing the catalyst or controlling the
geometry of the silyl ketene acetal. Only relatively simple
4.2. Anti-Selective Glycolate Aldol Reactions. On the basis
of the apanti transition structure (Figure 4), the observed reactivity
trend of silyl ketene acetals can be rationalized. As the alkyl
group of the ester is pointing toward the catalyst, the reaction
J. Org. Chem. Vol. 73, No. 12, 2008 4593

Denmark and Chung
FIGURE 4. Six open transition structures for glycolate aldol reactions with a bulky ester.
NaHCO3 solution (10 mL) and a satd aq KF solution (10 mL).
The biphasic mixture was stirred vigorously for 2 h at room
temperature. The mixture was filtered through a glass frit, and the
filtrate was transferred to a 125-mL separatory funnel where the
organic layer was separated. The aqueous layer was extracted with
CH2Cl2 (2 × 20 mL). The combined organic extracts were dried
over Na2SO4 (15 g), filtered, and concentrated in vacuo (23 °C, 30
mmHg). The residue was purified by column chromatography (18
mm diameter, hexane/EtOAc, 5/1 to 1/1) on silica gel (10 g) to
give 6da (275 mg, 87%) as a colorless oil. The syn/anti ratio was
determined to be 99/1 by 1H NMR (500 MHz) analysis of the crude
reaction mixture. Data for (2R,3S)-6da: 1H NMR (500 MHz, CDCl3)
δ 7.36-7.33 (m, 2 H, HC(12)), 7.31-7.22 (m, 8 H, HC(5), HC(6),
HC(7), HC(13), HC(14)), 4.79 (d, J ) 6.1, 1 H, HC(3)), 3.86 (d,
J ) 6.1, 1 H, HC(2)), 3.40 (s, 3 H, HC(8)), 2.98 (bs, 1 H, OH),
1.53 (s, 3 H, HC(10)), 1.52 (s, 3 H, HC(10)); 13C NMR (126 MHz,
CDCl3) δ 172.3 (C(1)), 143.8 (C(11)), 138.8 (C(4)), 128.1 (C(6)
or C(13)), 128.03 (C(6) or C(13)), 127.99 (C(7)), 127.4 (C(14)),
126.5 (C(5)), 126.1 (C(12)), 78.8 (C(9)), 77.4 (C(2)), 75.0 (C(3)),
51.5 (C(8)), 27.9 (C(10)), 27.1 (C(10)); IR (neat) ν 3490 (br), 3062
(w), 3031 (w), 2981 (m), 2950 (w), 1745 (s), 1496 (m), 1451 (m),
1435 (m), 1385 (m), 1368 (m), 1264 (m), 1197 (m), 1170 (m),
1151 (m), 1086 (s), 1076 (s), 1059 (s), 1028 (m), 983 (m), 914
(m), 879 (m), 767 (s), 722 (m), 700 (s); MS (ESI) 119.1 (86), 120.1
modification of the size of the alkyl groups on the silyl ketene
acetal was sufficient to reverse the stereochemical course. The
Lewis base catalyzed glycolate aldol reaction should provide
an attractive alternative for the synthesis of enantiomerically
enriched syn- or anti-1,2-diol units. The observed diastereose-
lectivity and reactivity could be rationalized by the analysis of
six open transition structures with the aid of computational
analysis.
Experimental Section
Representative Procedure for Syn-Selective Glycolate Al-
dol. Preparation of (2R,3S)-3-Hydroxy-2-(1-methyl-1-phe-
nylethoxy)-3-phenylpropanoic Acid Methyl Ester ((2R,3S)-
6da) (Table 2, Entry 1). To a flame-dried, 10-mL Schlenk flask
fitted with a magnetic stir bar, a thermocouple, a gas inlet tube,
and a septum were added (R,R)-1 (8.4 mg, 0.01 mmol, 0.01 equiv),
CH2Cl2 (5 mL), and benzaldehyde (101.6 µL, 1.0 mmol). The
solution was cooled to -78 °C (internal temperature) in a dry
ice-acetone bath. After diisopropylethylamine (17.4 µL, 0.1 mmol,
0.1 equiv) and SiCl4 (126 µL, 1.1 mmol, 1.1 equiv) were added to
the flask via syringe, a solution of 4d (336.5 mg, 1.2 mmol, 1.2
equiv) in CH2Cl2 (5 mL) was added dropwise via syringe over 15
min. The internal temperature was kept below -70 °C during the
addition of 4d. The reaction mixture was stirred for additional 15
min at -78 °C before a mixture of MeOH (1 mL), Et3N (1 mL),
and CH2Cl2 (5 mL) was added. The resulting solution was
transferred into a 125-mL Erlenmeyer flask containing a satd aq
(5), 136.1 (15), 179.1 (35), 214.1 (8), 332.2 (100), 337.1 (M+
+
Na, 58), 338.1 (6), 353.1 (23); HRMS calcd for C19H22O4Na
337.1416, found 337.1413; TLC Rf 0.27 (hexane/EtOAc, 4/1)
[UV(254)/KMnO4]; SFC (2R,3S)-6da, tR 7.61 min (96.6%); (2S,3R)-
4594 J. Org. Chem. Vol. 73, No. 12, 2008

Lewis Base ActiVation of Lewis Acids
6da, tR 8.95 min (3.4%) (Chiralpak AS, 125 bar, 40 °C, 1.6%
MeOH in CO2, 2.5 mL/min, 220 nm); [R]24D 42.2 (c ) 1.0, EtOH).
Anal. Calcd for C19H22O4 (314.38): C, 72.59; H, 7.05. Found: C,
72.59; H, 7.08.
22.6 (C(11)), 7.9 (C(10)), 7.8 (C(10)) IR (neat) ν 3474 (br), 3064
(w), 3033 (w), 2976 (m), 2941 (m), 2884 (m), 2829 (w), 1734 (s),
1495 (w), 1456 (m), 1376 (m), 1266 (m), 1196 (s), 1151 (m), 1124
(s), 1065 (m), 980 (m), 846 (m), 752 (m), 700 (s), 613 (m); MS:
(ESI) 179.1 (100), 180.1 (3), 197.1 (18), 298.2 (36), 303.2 (M+
+
Representative Procedure for Anti-Selective Glycolate Al-
dol. Preparation of (2S,3S)-3-Hydroxy-2-methoxy-3-phenylpro-
panoic Acid 1-Ethyl-1-methylpropyl Ester ((2S,3S)-6ha) (Table
4, Entry 1). Following the representative procedure above, (R,R)-1
(8.4 mg, 0.01 mmol, 0.01 equiv) was combined with diisopropy-
lethylamine (17.4 µL, 0.1 mmol, 0.1 equiv), benzaldehyde (101.6
µL, 1.0 mmol), SiCl4 (126 µL, 1.1 mmol, 1.1 equiv), and 4h (346.2
mg, 1.2 mmol, 1.2 equiv) to yield, after column chromatography
(30 mm diameter, hexane/EtOAc, 5/1) on silica gel (30 g), 6ha
(255 mg, 91%) as a colorless oil. The syn/anti ratio was determined
to be >1/99 by 1H NMR (500 MHz) analysis of the crude reaction
mixture. Data for (2S,3S)-6ha: 1H NMR (500 MHz, CDCl3) δ
7.41-7.40 (m, 2 H, HC(5)), 7.35-7.32 (m, 2 H, HC(6)), 7.29-7.26
(m, 1 H, HC(7)), 4.96 (d, J ) 5.6, 1 H, HC(3)), 3.89 (d, J ) 5.6,
1 H, HC(2)), 3.40 (s, 3 H, HC(12)), 2.96 (bs, 1 H, OH), 1.81-1.73
(m, 3 H, HC(9)), 1.60 (dq, J ) 13.9, 7.5, 1 H, HC(9)), 1.30 (s, 3
H, HC(11)), 0.77 (dd, J ) 7.5, 7.5, 3 H, HC(10)), 0.74 (dd, J )
7.5, 7.5, 3 H, HC(10)); 13C NMR (126 MHz, CDCl3) δ 169.4 (C(1)),
139.6 (C(4)), 128.2 (C(6)), 127.9 (C(7)), 126.8 (C(5)), 87.7 (C(8)),
84.6 (C(2)), 74.0 (C(3)), 58.8 (C(12)), 30.3 (C(9)), 30.2 (C(9)),
Na, 7); HRMS calcd for C16H24O4Na 303.1572, found 303.1559;
TLC Rf 0.27 (hexane/EtOAc, 5/1) [UV(254)/KMnO4]; SFC (2S,3S)-
6ha, tR 10.81 min (95.1%); (2R,3R)-6ha, tR 11.40 min (4.9%)
(Chiralpak AD, 125 bar, 40 °C, 2.0% MeOH in CO2, 2.0 mL/min,
220 nm); [R]24 -5.6 (c ) 4.0, EtOH). Anal. Calcd for C16H24O4
D
(280.36): C, 68.54; H, 8.63. Found: C, 68.35; H, 8.72.
Acknowledgment. We are grateful to the National Science
Foundation for generous financial support (NSF CHE 0414440
and CHE 0717989).
Supporting Information Available: Full characterization of
all enol ethers and aldol products along with representative
procedures for the addition reactions and configurational as-
signments as well as atomic coordinates for the calculated
transition structure. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO8006539
J. Org. Chem. Vol. 73, No. 12, 2008 4595