Lewis base catalyzed aldol additions of chiral trichlorosilyl enolates and silyl enol ethers

Article abstract of DOI:10.1021/jo051930+

The consequences of double diastereodifferentiation in chiral Lewis base catalyzed aldol additions using chiral enoxysilanes derived from lactate, 3-hydroxyisobutyrate, and 3-hydroxybutyrate have been investigated. Trichlorosilyl enolates derived from the

Full text of DOI:10.1021/jo051930+

Lewis Base Catalyzed Aldol Additions of Chiral Trichlorosilyl
Enolates and Silyl Enol Ethers
Scott E. Denmark,* Shinji Fujimori, and Son M. Pham
Department of Chemistry, Roger Adams Laboratory, University of Illinois, Urbana, Illinois 61801
denmark@scs.uiuc.edu
Received September 15, 2005
The consequences of double diastereodifferentiation in chiral Lewis base catalyzed aldol additions
using chiral enoxysilanes derived from lactate, 3-hydroxyisobutyrate, and 3-hydroxybutyrate have
been investigated. Trichlorosilyl enolates derived from the chiral methyl and ethyl ketones were
subjected to aldolization in the presence of phosphoramides, and the intrinsic selectivity of these
enolates and the external stereoinduction from chiral catalyst were studied. In the reactions with
the lactate derived enolate, the strong internal stereoinduction dominated the stereochemical
outcome of the aldol addition. For the 3-hydroxyisobutyrate- and 3-hydroxybutyrate derived enolates,
the catalyst-controlled diastereoselectivities were observed, and the resident stereogenic centers
exerted marginal influence. The corresponding trimethylsilyl enol ethers were employed in SiCl4/
bisphosphoramide catalyzed aldol additions, and the effect of double diastereodifferentiation was
also investigated. The overall diastereoselection of the process was again controlled by the strong
external influence of the catalyst.
Introduction
stereogenic center(s) in the substrate. A careful analysis
of the matching/mismatching interaction between the two
modes of stereoinduction is required for installation of
Enantioselective aldol additions are among the most
useful synthetic transformations in organic synthesis.
The development of highly diastereo- and enantioselec-
tive variants has been amply documented in numerous
recent reviews.1 High stereoselectivities can be achieved
using chiral auxiliaries2 and chiral catalysts.3 When these
methods are used in conjunction with a chiral substrate,
these reactions are called double diastereodifferentiat-
ing aldol additions. In these cases, two different modes
of stereoinduction are operating in the aldolization
process: induction from catalyst or auxiliary and from
chiral substrate. The concept of double diastereodiffer-
entiation has been studied in detail, and several authori-
tative reviews have appeared.4 The presence of stereo-
genic centers in substrates can complicate the stereo-
selective processes in the construction of complex mol-
ecules because the selectivity of the reaction is affected
not only by the catalyst or auxiliary but also by the
(1) (a) Evans, D. A.; Nelson, J. V.; Taber, T. R. In Topics in
Stereochemistry; Eliel, E. L., Wilen, S. H., Eds.; Wiley-Interscience:
New York, 1982; Vol. 13; pp 1-115. (b) Heathcock, C. H. In Compre-
hensive Carbanion Chemistry; Buncel, E., Durst, T., Eds.; Elsevier:
New York, 1984; Vol. 5B, pp 177-237. (c) Mukaiyama, T. Org. React.
1982, 28, 203-331. (d) Braun, M. Angew. Chem., Int. Ed. Engl. 1987,
26, 24-37. (e) Mukaiyama, T.; Kobayashi, S. Org. React. 1994, 46,
1-103. (f) Norcross, R.; Paterson, I. Chem. Rev. 1995, 95, 2041-2114.
(g) Carreira, E. M. In Modern Carbonyl Chemistry; Otera, J., Ed.;
Wiley-VCH: Weinheim, 2000; Chapter 8. (h) Paterson, I.; Cowden, C.
J.; Wallace, D. J. In Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-
VCH: Weinheim, 2000; Chapter 9. (i) Carreira, E. M. In Comprehensive
Asymmetric Catalysis, Vols. I-III; Jacobsen, E. N., Pfaltz, A., Yama-
moto, H., Eds.; Springer-Verlag: Heidelberg, 1999. Chapter 29. (j)
Carreira, E. M. In Catalytic Asymmetric Synthesis, 2nd ed.; Ojima, I.,
Ed.; Wiley-VCH: Weinheim, 2000; Chapter 8B2. (k) Braun, M. In
Stereoselective Synthesis, Methods of Organic Chemistry (Houben-Weyl),
ed. E21; Helmchen, G., Hoffman, R., Mulzer, J., Schaumann, E., Eds.;
Thieme: Stuttgart, 1996; Vol. 3; pp 1603-1735. (l) Modern Aldol
Reactions; Mahrwald, R., Ed.; Wiley-VCH: Weinheim, 2004; Vols. 1
and 2.
10.1021/jo051930+ CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/18/2005
J. Org. Chem. 2005, 70, 10823-10840
10823

Denmark et al.
new stereogenic centers in a predictable and controlled
manner.
aldehydes, there is a third stereoselection process that
is controlled by resident stereogenic centers. The dias-
tereoselection resulting from the influence of the stereo-
genic centers in either of the reactants is referred as
internal diastereoselection. For example, when a chiral
catalyst is used in conjunction with a chiral enolate, there
is a possibility of double diastereodifferentiation. In a
matched case wherein the sense of external stereoinduc-
tion coincides with the internal stereoinduction, the
diastereoselectivity of the reaction can be considerably
enhanced.
Over the past 10 years, catalytic enantioselective aldol
additions using chiral Lewis bases have been developed
in these laboratories.7 In the presence of a chiral phos-
phoramide, trichlorosilyl enolates derived from various
carbonyl compounds undergo stereoselective aldol addi-
tions. More recently, it was found that aldol additions of
trialkylsilyl enol ethers and ketene acetals are also
catalyzed by chiral phosphoramides in the presence of
silicon tetrachloride. The scope of nucleophile in these
reactions has been expanded to include chiral silyl enol
ethers.
When discussing double diastereo-differentiating aldol
additions, the following nomenclature allows the different
kinds of stereoinduction to be clearly distinguished.5
There are three types of stereoselection processes associ-
ated with typical aldol additions (Scheme 1). The first is
the relative diastereoselection, which corresponds to the
relative topicity (like or unlike)6 of the two reacting faces
(enolate and carbonyl group). In highly organized aldol
additions, this is often interpreted in terms of the chair/
boat selectivity in the transition structure. The relative
diastereoselection only pertains to R-substituted enolates
since the term refers to the relative configuration of the
two substituents (like or unlike) at the newly created
stereogenic centers. The other is the absolute stereose-
lection (external stereoselection) determined by a chiral
reagent. This term is used to describe the enantiofacial
outcome at the newly created stereogenic centers. In the
addition of chiral enolates or the addition to chiral
SCHEME 1
In this report, our investigations on the use of chiral
nucleophiles for asymmetric aldol additions is fully
described.8 The primary objective of this study was to
evaluate the effect of a stereogenic center on the stere-
ochemical outcome of the aldol addition, including the
effect of double diastereodifferentiation using chiral
phosphoramides. Initial studies would involve the aldol
additions of these chiral enol ethers to benzaldehyde to
establish the reaction conditions, and then the substrate
scope with respect to aldehyde would be investigated.
Three classes of chiral silyl enol ethers were chosen to
investigate the effect of the resident stereogenic center
in the context of Lewis base catalyzed aldol addition
(Chart 1). Two of these enolates bear heteroatom-based
stereogenic centers either at the R- or â-carbons on the
spectator side (I and II), and one bears a carbon-based
stereogenic center on the spectator side (III). These types
of nucleophiles were chosen because the aldol products
derived from these enolates and enol ethers resemble the
1,2,4- and 1,3,5-triol structural motifs that are commonly
found in natural products.
CHART 1
(2) For example: Evans, D. A.; Bartroli, J.; Shih, T. L. J. Am. Chem.
Soc. 1981, 103, 2127-2129.
(3) For examples: (a) Lewis Acids in Organic Synthesis; Yamamoto,
H., Ed.; Wiley-VCH: Weinheim, 2001; Vols. 1 and 2. (b) Santelli, M.;
Pons, J.-M. Lewis Acids and Selectivity in Organic Synthesis; CRC:
Boca Raton, 1996. (c) Kobayashi, S.; Uchiro, H.; Mukaiyama, T.
Tetrahedron 1993, 49, 1761-1772. (d) Nelson, S. G. Tetrahedron
Asymmetry 1998, 9, 357-389. (e) Kobayashi, S.; Fujishita, Y.; Mu-
kaiyama, T. Chem. Lett. 1990, 1455-1458. (f) Evans, D. A.; MacMillan,
D. W. C.; Campos, K. R. J. Am. Chem. Soc. 1997, 119, 10859-10860.
(g) Kobayashi, S.; Horibe, M.; Hachiya, I. Tetrahedron Lett. 1995, 36,
3173-3176. (h) Yamashita, Y.; Ishitani, H.; Shimizu, H.; Kobayashi,
S. J. Am. Chem. Soc. 2002, 124, 3292-3302. (i) Yamada, Y. M. A.;
Yoshikawa, N.; Sasai, H.; Shibasaki, M. Angew. Chem., Int. Ed. Engl.
1997, 36, 1871-1873. (j) Yoshikawa, N.; Yamada, Y. M. A.; Das, J.;
Sasai, H.; Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 4168-4178. (k)
Shibasaki, M.; Matsunaga, S.; Kumagai, N. In Modern Aldol Reactions;
Mahrwald, R., Ed.; Wiley-VCH: Weinheim, 2004; Vol. 2. Chapter 6.
(4) (a) Masamune, S.; Choy, W.; Petersen, J. S.; Sita, L. R. Angew.
Chem., Int. Ed. Engl. 1985, 24, 1-76. (b) Kolodiazhnyi, O. I. Tetra-
hedron 2003, 59, 5953-6018.
Background
As mentioned in the introduction, two types of Lewis
base catalyzed aldol reactions have been developed that
differ by the nucleophilic addition partner. The first of
these methods employs trichlorosilyl enolates. These
(7) (a) Denmark, S. E.; Stavenger, R. A. Acc. Chem. Res. 2000, 33,
432-440. (b) Denmark, S. E.; Fujimori, S. In Modern Aldol Reactions;
Mahrwald, R., Ed.; Wiley-VCH: Weinheim, 2004; Vol. 2. Chapter 7.
(8) For preliminary communications, see: (a) Denmark, S. E.;
Fujimori, S. Synlett 2000, 1024-1029. (b) Denmark, S. E.; Pham, S.
M. Org. Lett. 2001, 3, 2201-2204. (c) Denmark, S. E.; Fujimori, S.
Org. Lett. 2002, 4, 3477-3480. (d) Denmark, S. E.; Fujimori, S. Org.
Lett. 2002, 4, 3473-3476.
(5) Denmark, S. E.; Almstead, N. G. In Modern Carbonyl Chemistry;
Otera, J., Ed.; Wiley-VCH: Weinheim, 2000; Chapter 10.
(6) Seebach, D.; Prelog, V. Angew. Chem. 1982, 21, 654-660.
10824 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
species undergo highly stereoselective aldol additions in
the presence of a chiral phosphoramide such as (S,S)-1
(Scheme 2).9 Trichlorosilyl enolates derived from ac-
etates,10 methyl ketones,9 ethyl ketones,11 cyclic ketones,12
and aldehydes13 are competent and selective nucleophiles
in this aldolization process.
is significantly higher when (R,R)-1 is used as the
catalyst compared to (S,S)-1. This observation indicates
that there is matched and mismatched relationship4
between the stereoinduction by the catalyst and the
resident stereogenic center. In the matched case, there
is a strong enhancement of the syn selectivity because
the sense of stereoinduction by the catalyst and that of
the chiral enolate are the same. On the other hand, the
use of (S,S)-1 gives rise to an eroded syn selectivity
because the stereoselection caused by the catalyst is
opposite to the effect of resident stereogenic center. The
resulting syn selectivity in this case is largely controlled
by the strong stereoinduction from the R-silyloxy stereo-
genic center in the chiral enolate.
SCHEME 2
The second type of Lewis base catalysis involves
activation of weak Lewis acids for Mukaiyama-type aldol
additions of trialkylsilyl enol ethers and ketene acetals.14
These nucleophiles undergo highly stereoselective aldol
addition to aromatic and conjugated aldehydes in the
presence of SiCl4 and chiral bisphosphoramide 2 as the
catalyst (Scheme 3). The aldol addition of propanoate
derived ketene acetals shows excellent anti (relative)
diastereoselectivity and enantioselectivity.14d This cata-
lyst system remains as one of the very few anti-selective
catalytic, enantioselective aldol additions.15
SCHEME 4
In the following sections, the aldol additions of chiral
nucleophiles catalyzed by chiral Lewis bases are de-
scribed. The results are organized as follows: (1) prepa-
rations of these chiral silyl nucleophiles, (2) the aldol
additions of the methyl ketone derived nucleophiles, and
(3) the aldol additions of the ethyl ketone derived
nucleophiles. In the description of aldol additions, the
reactions of trichlorosilyl enolates are presented first, and
then the additions of trimethylsilyl enol ethers are
described.
SCHEME 3
Results
1. Preparation of Chiral Silyl Enol Ethers. The
chiral methyl ketones 7 and 10 were prepared from
methyl (S)-3-hydroxyisobutyrate and from methyl (S)-3-
hydroxybutyrate, respectively.8 The TBS ether was cho-
sen as the protective group for the hydroxyl group on the
basis of previous studies.16 The site-selective enolizations
of the methyl ketones were accomplished by deprotona-
tion with lithium diisopropylamide followed by the reac-
tion with TMSCl (Scheme 5). The corresponding TMS
enol ethers 8 and 11 were obtained in excellent yields.
The conversion of the TMS enol ethers to the trichlorosi-
In the context of double diastereodifferentiation, the
lactate derived trichlorosilyl enolates were among the
first to be examined (Scheme 4).16 The aldol addition of
5 is efficiently catalyzed by phosphoramides (R,R)- and
(S,S)-1, and in both cases the 1,4-syn diastereomer is
obtained as the major product. The observed selectivity
(9) Denmark, S. E.; Stavenger, R. A. J. Am. Chem. Soc. 2000, 122,
8837-8847.
(10) (a) Denmark, S. E.; Winter, S. B. D.; Su, X.; Wong, K.-T. J.
Am. Chem. Soc. 1996, 118, 7404-7405. (b) Denmark, S. E.; Fan, Y. J.
Am. Chem. Soc. 2002, 124, 4233-4235.
SCHEME 5
(11) Denmark, S. E.; Pham, S. M. J. Org. Chem. 2003, 68, 5045-
5055.
(12) Denmark, S. E.; Stavenger, R. A.; Wong, K.-T. Tetrahedron
1998, 54, 10389-10402.
(13) (a) Denmark, S. E.; Ghosh, S. K. Angew. Chem., Int. Ed. 2001,
40, 4759-4762. (b) Denmark, S. E.; Bui, T. Proc. Nat. Acad. Sci. 2004,
101, 5439-5444.
(14) (a) Denmark, S. E.; Wynn, T.; Beutner, G. L. J. Am. Chem. Soc.
2002, 124, 13405-13407. (b) Denmark, S. E.; Heemstra, J. R., Jr. Org.
Lett. 2003, 5, 2303-2306. (c) Denmark, S. E.; Beutner, G. L. J. Am.
Chem. Soc. 2003, 125, 7800-7801. (d) Denmark, S. E.; Beutner, G.
B.; Wynn, T.; Eastgate, M. D. J. Am. Chem. Soc. 2005, 127, 3774-
3789.
(15) For examples of anti-selective catalytic aldol additions, see: (a)
Evans, D. A.; MacMillan, D. W. C.; Campos, K. R. J. Am. Chem. Soc.
1997, 119, 10859-10860. (b) Kobayashi, S.; Horibe, M.; Hachiya, I.
Tetrahedron Lett. 1995, 36, 3173-3176. (c) Yamashita, Y.; Ishitani,
H.; Shimizu, H.; Kobayashi, S. J. Am. Chem. Soc. 2002, 124, 3292-
3302.
(16) Denmark, S. E.; Stavenger, R. A. J. Org. Chem. 1998, 63, 9524-
9527.
J. Org. Chem, Vol. 70, No. 26, 2005 10825

Denmark et al.
lyl enolates has been studied in detail.17 Either Hg(II)
or Pd(II) acetate efficiently catalyzed the transsilylation
of 8 and 11 with SiCl4 to afford the trichlorosilyl enolates
9 and 12 in good yields.
SCHEME 8
Geometrically defined TMS enol ethers from the chiral
ethyl ketones were prepared under the following condi-
tions. The Z TMS enol ethers were generated by a
combination of TMSOTf/Et3N18 or TMSCl/DBU19 (Scheme
6), whereas the E TMS enol ethers were formed by the
use of lithium tert-butyltritylamine (LiTBTA),20 affording
(E)-17 and (E)-22 (Scheme 7).
SCHEME 6
Geometrically defined trichlorosilyl enolates are pre-
pared more reliably by an alternative two-step procedure
that involves lithium enolate formation followed by
silylation with silicon tetrachloride (Scheme 9).13 By the
use of this method, both (E)- and (Z)-trichlorosilyl eno-
lates can be prepared without erosion of the E/Z ratio
of the starting TMS enol ether. The resulting enolates
were purified by bulb-to-bulb distillation; however, vari-
able amounts of bisenoxydichlorosilane species were
observed in the isolated trichlorosilyl enolates.
SCHEME 9
SCHEME 7
The preparation of geometrically defined trichlorosilyl
enolates from achiral ethyl ketones has been studied.11
It has been shown that (Z)-trichlorosilyl enolates can be
prepared selectively using the Hg(II)- or Pd(II)-catalyzed
transsilylation. However, the E/Z-selectivity of the trans-
silylation is dependent on the size of the spectator
group: larger spectator groups show higher Z-selectivity.
The Hg(II)-catalyzed transsilylations gave variable re-
sults for the chiral ethyl ketone enolates (Scheme 8).
The lactate derived TMS enol ether 14 provided enolate
(Z)-23 with high selectivity. However, the transsilylation
of either (Z)- or (E)-17 under similar conditions resulted
in formation of 1/1 mixture of 24.
2. Aldol Additions of Methyl Ketone Derived
Chiral Trichlorosilyl Enolates. 2.1. Addition of
Trichlorosilyl Enolate 9. Optimization of the aldol
addition of 9 with benzaldehyde involved a survey of the
reaction concentration, catalyst structure, and loading
(Table 1). The chiral phosphoramide (R,R)-1 effectively
promoted the aldol addition at 5 mol % loading and a
0.5 M reaction concentration to afford syn-28a as the
major isomer.21 The increasing the catalyst loading
(21) The configurational assignment of the major diastereomer as
syn-28a was confirmed by single-crystal X-ray analysis of the diol 31
obtained by deprotection of TBS ether with PPTS in ethanol. (a)
Martin, V. A.; Murray, D. H.; Pratt, N. E.; Zhao, Y.; Albizati, K. F. J.
Am. Chem. Soc. 1990, 112, 6965-6968. Recrystallization of the crude
diol provided a crystal suitable for X-ray analysis. (b) The crystal-
lographic coordinates of 31 have been deposited with the Cambridge
Crystallographic Data Centre, deposition no. CCDC 270359. These data
can be obtained free of charge via www.ccdc.cam.ac.uk/conts/retriev-
ing.html (or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, U.K.; fax (+44) 1223-336-033; or
deposit@ccdc.cam.ac.uk).
(17) Denmark, S. E.; Stavenger, R. A.; Winter, S. B. D.; Wong, K.-
T.; Barsanti, P. A. J. Org. Chem. 1998, 63, 9517-9523.
(18) (a) Trost, B. M.; Urabe, H. J. Org. Chem. 1990, 55, 3982-3984.
(b) Vorbruggen, H.; Krolikiewicz, K.; Bennua, B. Chem. Ber. 1981, 114,
1234-1235. (c) Corey, E. J.; Cho, H.; Rucker, C.; Hua, D. H. Tetrahe-
dron Lett. 1981, 22, 3455-3458.
(19) Taniguchi, Y.; Inanaga, J.; Yamaguchi, M. Bull. Chem. Soc. Jpn.
1981, 54, 3229-3232.
(20) Busch-Petersen, J.; Corey, E. J. Tetrahedron Lett. 2000, 41,
2515-2518.
10826 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
CHART 2
showed no influence on the diastereoselectivity (entries
2-4). The enantiomeric catalyst (S,S)-1 was equally
effective in promoting the aldol addition (5 mol % loading
at 0.5 M) to afford anti-28a as the major product (entry
8). Lower reaction concentration can be used if the
catalyst concentration is increased.
In the presence of an achiral phosphoramide, intrinsic,
internal diastereoselectivity in the catalyzed reaction was
investigated. The N,N-dimethyl phosphoramide 29 pro-
vided the syn-28a as the major product with modest
diastereoselectivity. The reaction using N,N-diphenyl
phosphoramide 30 was significantly slower and only 31%
yield of 28a was obtained after 12 h.
Thus, it was found that (R,R)-1 yielded syn-28a selec-
tively (syn/anti, 10/1), whereas (S,S)-1 afforded anti-28a
with attenuated selectivity (syn/anti, 1/6). In the presence
of achiral phosphoramide 29, the reaction weakly favored
the syn product, indicating that there is a mild inherent
selectivity toward the syn product by the resident ste-
reogenic center (Table 1, entry 10). However, the stereo-
selection in this aldol reaction was dominated by the
catalyst configuration since there was a switch in the
sense of asymmetric induction.
improved. In general, the aldol addition of in situ
generated 9 was efficiently catalyzed by phosphoramide
1 affording 28a in good yields except for aliphatic
aldehydes (entries 9 and 10).
The effect of the resident stereogenic center was again
examined using achiral phosphoramide 29 for the in situ
generated trichlorosilyl enolate 9. The addition showed
a slight preference to the syn isomer consistent with the
previous experiment (Table 2, entry 12).
TABLE 1. Optimization of the Aldol Addition of 9 with
Phosphoramides 1, 29, and 30
The diastereoselectivities were dependent on aldehyde
structure, but the selectivities were always higher when
(R,R)-1 was used as the catalyst.22 The selectivity seemed
to depend on the substitution pattern of the aldehyde
such that aldehydes containing R-branching generally
gave higher diastereoselectivity. For example, the addi-
tions to pivaldehyde and tiglic aldehyde afforded the
highest 1,4-syn selectivity when (R,R)-1 was used (Table
2, entries 7 and 11). The lowest syn selectivity was
obtained with the acetylenic aldehyde (entry 8) presum-
ably due to the absence of sterically demanding groups
leading to less steric differentiation of the π-faces. The
yields for the aliphatic aldehydes were significantly lower
because these reactions did not go completion, though the
entry catalyst concn, M cat., mol % time, h yield, %a syn/antib
1
2
3
4
5
6
7
8
9
(R,R)-1
(R,R)-1
(R,R)-1
(R,R)-1
(S,S)-1
(S,S)-1
(S,S)-1
(S,S)-1
29
0.1
0.5
0.5
0.5
0.1
0.2
0.2
0.5
0.1
0.1
5
5
2
2
43c
79
14/1
10/1
10/1
10/1
1/9
10
15
15
10
15
5
1
73
1
71
3
57c
81
2
1/6
2
72c
76c
51c
31c
1/6
2
1/6
(22) The absolute configuration of the aldol products was assigned
by analogy to the benzaldehyde aldol product. However, to ensure the
consistency of stereoinduction from the catalyst system across the
various aldehyde structures, the configuration of the â-hydroxy ketone
obtained from pivaldehyde (syn-28k) was assigned by chemical cor-
relation to the known acid 34 The site-selective Baeyer-Villiger
oxidation of the ketone was accomplished using conditions developed
by Shibasaki et al. to provide the ester 33 (Gottlich, R.; Yamakoshi,
K.; Sasai, H.; Shibasaki, M. Synlett 1997, 971-973). The ester was
hydrolyzed to the corresponding acid in good yield, the optical rotation
of which compared well to the literature value of S-34. (Devant, R.;
Braun, M. Chem. Ber. 1986, 119, 2191-2207). Thus, the sense of
stereoinduction with (R,R)-1 was consistent between aldol additions
to benzaldehyde and to pivaldehyde.
5
15
5
5/1
10 30
12
2/1
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC (OD; 175 bar, 2.5 mL/min, 1.7% MeOH).
c Incomplete reaction.
Next, the generality of this aldol addition with other
aldehydes was investigated. For this study, the in situ
generation of 9 was employed to further demonstrate the
practicality of this aldol addition. For their transsilylation
1 mol % of Hg(OAc)2 was employed for the in situ
formation of trichlorosilyl ether 9. The standard condi-
tions for surveying aldehydes employed 10 mol % of the
catalyst at 0.5 M overall reaction concentration (Table
2). The aldehydes surveyed include aromatic, olefinic,
acetylenic and aliphatic with various substitution pat-
terns (Chart 2).
The in situ method gave superior results compared to
those obtained with isolated 9 (Table 2, entry 1). The
overall yields of the aldol products from the TMS enol
ether were significantly improved, and the diastereose-
lectivities of the aldol additions were also noticeably
J. Org. Chem, Vol. 70, No. 26, 2005 10827

Denmark et al.
TABLE 2. Aldol Addition of in Situ Generated Trichlorosilyl Enolate 9 to Aldehydes
(R,R)-1
(S,S)-1
entry
aldehyde R
phenyl
product
time, h
yield, %a
dr (syn/anti)b
yield, %a
dr (syn/anti)a
1
2
1a
1b
1c
1d
1e
1f
2.5
2
80
85
81
84
85
81
78
83
34
47
73
83
19.0/1
15.7/1
8.00/1
10.1/1
4.88/1c
15.7/1c
24.0/1c
3.35/1
10.1/1
15.7/1c
27.9/1e
4.88/1
75
83
82
80
83
84
75
82
22
31
78
1/7.33
1/8.09
1/4.26
1/6.69
1/2.87c
1/8.09c
1/4.56c
1/1.32
1/2.45
1/4.00c
1/6.51e
1-naphthyl
(E)-cinnamyl
(E)-2-Me-cinnamyl
(E)-crotyl
3
7
4
4
5
4
6
CH2dCMe
MeCHdCMe
PhCtC-
2
7
1g
1h
1i
4
8
5
9
PhCH2CH2
cyclohexyl
t-butyl
10d
10d
4
10
11
12f
1j
1k
phenyl
1a
2.5
a Yield of analytically pure material. b Determined by CSP-SFC. c Determined by CSP-SFC on the corresponding benzoate. d Incomplete
reaction. e Determined by 1H NMR analysis in d6-benzene. f 10 mol % of 29 was used as catalyst.
selectivities were good for the matched cases (entries 9
and 10). The low conversion in these cases was believed
to arise from the formation of an unreactive R-chloro silyl
ether under these reaction conditions.7
The addition of the methyl ketone enolates can be
performed using the in situ generated enolate for opera-
tional simplicity (Table 3). This method not only avoided
handling of sensitive trichlorosilyl enolates but also
provided improved selectivities for the additions of 11 as
compared to those of isolated 12 (Table 3, entries 1 and
2). The stereochemical course of the reaction seems to
be controlled only by the catalyst because similar selec-
tivities for both diastereomers were obtained using either
enantiomer of 1. The addition of 36 provided good yields
of the aldol products; however, the selectivity was
significantly attenuated compared to the aldol additions
of 9 (Table 3, entries 3 and 4).
2.2. Addition of Trichlorosilyl Enolate 12. In the
previous section, the effect of an R-stereogenic center
bearing carbon substituents on the stereochemical course
of the aldol addition of trichlorosilyl enolate was studied.
The effect of a remote stereogenic center on diastereo-
selection of the aldol addition is also worthy of investiga-
tion because the resulting aldol product resembles the
1,3-polyol fragment present in numerous natural prod-
ucts.23 In the aldol addition of boron enolates, it has been
demonstrated that a â-oxygen bearing stereogenic center
can strongly influence the stereochemical course of the
reaction.23
TABLE 3. Aldol Addition of In Situ Generated
Trichlorosilyl Enolates to Benzaldehyde
The aldol addition of the methyl ketone enolate 12 to
benzaldehyde was examined under catalysis by the chiral
and achiral phosphoramides 1 and 29 (Scheme 10). The
use of 10 mol % of the catalyst effectively promoted the
addition at 0.5 M reaction concentration. The intrinsic
(substrate-induced) selectivity determined using 29 was
almost negligible, indicating that the â-stereogenic center
does not exert significant stereoinduction under phosphor-
amide catalysis. Unfortunately, with either enantiomer
of the chiral catalyst, only marginal diastereoselectivities
were obtained. These results are perplexing considering
that the phosphoramide 1 shows high selectivities in the
aldol additions with other methyl ketone enolates.9
entry TMS enol ether catalyst yield, %a dr (syn/anti)b
1
2
3
4
11
11
36
36
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
61
58
89
86
1/5.5
4.4/1
1/1.8
2.3/1
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC.
SCHEME 10
3. Aldol Additions of Methyl Ketone Derived
Chiral TMS Enol Ethers. 3.1. Aldol Addition of 8.
To examine the effect of the resident stereogenic center
on the stereochemical course of the reaction, the aldol
addition of chiral TMS enol ether 8 was investigated
using chiral bisphosphoramide 2 as the catalyst. The
initial optimization surveyed the catalyst structure, effect
of additives and reaction concentration (Table 4).
(23) (a) Paterson, I.; Gibson, K. R.; Oballa, R. M. Tetrahedron Lett.
1996, 37, 8585-8588. (b) Paterson, I.; Collett, L. A. Tetrahedron Lett.
2001, 42, 1187-1191. (c) Paterson, I.; Oballa, R. M.; Norcross, R. D.
Tetrahedron Lett. 1996, 37, 8581-8584. (d) Evans, D. A.; Coleman, P.
J.; Cote, B. J. Org. Chem. 1997, 62, 788-789. (e) Evans, D. A.; Cote,
B.; Coleman, P. J.; Connell, B. T. J. Am. Chem. Soc. 2003, 125, 10893-
10898.
With (R,R)-2, the addition of 8 to benzaldehyde pro-
vided syn-28a in excellent yield and diastereoselectivity.
On the other hand, when (S,S)-2 was used as catalyst,
anti-28a was obtained exclusively. The monomeric phos-
phoramide (R,R)-1 was not as catalytically active as
10828 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
TABLE 4. Aldol Addition of 8 to Benzaldehyde
(entries 3 and 4). On the other hand, addition to 2-naph-
thaldehyde (entry 5) and cinnamaldehyde (entries 6 and
7) proceeded smoothly to provide the aldol products in
high yield and good diastereoselectivity. The addition to
3-phenylpropargyl aldehyde showed a significant de-
crease in the selectivity, although the yield was good
(entries 8 and 9). Unfortunately, the aldol addition to
crotonaldehyde was found to be problematic due to the
formation of side products. The desired aldol products
could only be obtained in low yield (albeit high selectivity)
after chromatographic purification. The addition to ali-
phatic and R-substituted olefinic and R-substituted aro-
matic aldehydes (entries 12-14) provided little or no
product.
3.2. Aldol Addition of 11. The initial study was
carried out using benzaldehyde as the substrate (Scheme
11). In contrast to the results from aldol addition of
trichlorosilyl enolate 12, the aldol addition of 11 provided
aldol product 35a in good yield in the presence of 2 and
SiCl4. In addition, the selectivity observed was signifi-
cantly higher than the selectivity obtained from addition
of 12. The aldol addition of 11 to benzaldehyde catalyzed
by (R,R)-2 provided 1,5-anti 35a almost exclusively.24 The
1,5-syn diastereomer of 35a was obtained as the major
product when enantiomeric catalyst (S,S)-2 was used.
yield,
%
dr,
entry
catalyst
additive
syn/antib
1
2
3
4
5 mol % (R,R)-2 none
88
90
58
92
24/1
1/>99
1/3
5 mol % (S,S)-2
none
10 mol % (R,R)-1 none
5 mol % (R,R)-2 tri-tert-butylpyridine
(20 mol %)
5 mol % (R,R)-2 i-Pr2EtN (20 mol %)
5 mol % (R,R)-2 none
29/1
5
90
89
23/1
25/1
6c
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC. c Reaction was run at 1.0 M concentration.
bisphosphoramide, providing the aldol product in modest
yield and selectivity. The introduction of additives such
as tertiary amines and tetraalkylammonium salts has
shown beneficial effects in the related studies. The
addition of amines to scavenge adventitious acids helped
to increase the yields for the additions of achiral TMS
enol ethers. Indeed, in the additions of 8, the yield and
selectivities were increased slightly by the addition of 20
mol % of the amines (entries 4 and 5). Tri-tert-butylpy-
ridine provided optimal results; however, diisopropyl-
ethylamine was chosen as an additive because of its lower
cost. Interestingly, the reaction at higher concentration
did not improve the yield or the selectivity (entry 6).
SCHEME 11
TABLE 5. Aldol Addition of 8 to Various Aldehydes
Catalyzed by Bisphosphoramide 2
Because the diastereoselectivities were similar for both
(R,R)-2 and (S,S)-2 catalyzed reactions, the inherent
selectivity is almost nonexistent and the diastereoselec-
tivity is primarily controlled by the catalyst configuration.
This result is consistent with previous findings.8b
The high yields and selectivities observed for the
addition of 11 to benzaldehyde prompted a survey of the
aldehyde scope for this aldol addition (Table 6). Aromatic
and conjugated aldehydes afforded high yields of aldol
products with modest to excellent diastereoselectivities
(entries 3-6). In most cases, (S,S)-2 provided the 1,5-
syn aldol products, whereas 1,5-anti aldol products were
obtained using (R,R)-2 as the catalyst. For cinnamalde-
hyde and naphthaldehyde, 1,5-anti selectivity with (S,S)-2
was higher than that with the 1,5-syn stereoinduction
with (R,R)-2. The addition to naphthaldehyde was very
slow, but aldol products were obtained in good yields and
high diastereoselectivities. The reaction with 3-phenyl-
propargyl aldehyde was completely unselective in both
cases, providing a nearly 1/1 mixture of diastereomers
time, yield,
a
entry
aldehyde, R
product catalyst
h
%
syn/antib
1
2
3
4
5
6
7
8
9
phenyl
phenyl
28a
28a
(R,R)-2
(S,S)-2
4
91
94
85
83
77
85
83
74
72
23
27
22
nr
nr
24/1
1/49
19/1
1/43
99/1
32/1
1/34
1/1.5
1/4
4
1-naphthyl
1-naphthyl
2-naphthyl
(E)-cinnamyl
(E)-cinnamyl
PhCtC-
28b (R,R)-2 24
28b (S,S)-2
24
3
28l
28c
28c
(R,R)-2
(R,R)-2
(S,S)-2
5
5
28h (R,R)-2
28h (S,S)-2
6
PhCtC-
6
10 (E)-crotyl
11 (E)-crotyl
28e
28e
(R,R)-2
(S,S)-2
4
12/1
1/>19
4/1
4
12 (E)-2-Me-cinnamyl 28d (R,R)-2 24
13 PhCH2CH2
14 2-tolyl
28i
(R,R)-2 24
28m (R,R)-2
4
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC.
(24) The configuration of the aldol product was confirmed by single-
crystal X-ray analysis of the corresponding diol anti-38. The TBS group
was removed from anti-35a under acidic conditions to give diol anti-
38, which provided a crystal suitable for X-ray analysis from refluxing
acetone/hexanes. The crystallographic coordinates of anti-38 have been
deposited with the Cambridge Crystallographic Data Centre, deposition
no. CCDC 270165. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
U.K.; fax (+44) 1223-336-033; or deposit@ccdc.cam.ac.uk).
The high yields and selectivities obtained in the
addition of 8 to benzaldehyde prompted a study of the
scope of aldehyde structure in this reaction (Table 5). The
reaction with 1-naphthaldehyde showed a significant
decrease in the reaction rate; however, the reaction
proceeded to completion with increased reaction time
J. Org. Chem, Vol. 70, No. 26, 2005 10829

Denmark et al.
TABLE 6. Aldol Addition of 11 to Various Aldehydes
N-phenyl substituents on the 1,3,2-diazaphospholidine
ring afforded attenuated internal selectivities (entries 3
and 6).
TABLE 7. Stereoselective Aldol Additions of (Z)-23 to
Benzaldehyde Using Various Phosphoramides
time, yield,
a
entry
aldehyde, R
product catalyst
h
%
syn/antib
1
2
3
4
5
6
7
8
9
phenyl
35a (R,R)-2
35a (S,S)-2
5
92
94
74
79
88
90
74
85
nr
tr
1/36
37/1
phenyl
5
1-naphthyl
1-naphthyl
(E)-cinnamyl
(E)-cinnamyl
PhCtC
35b (R,R)-2 12
1/21
35b (S,S)-2
35c (R,R)-2
35c (S,S)-2
35h (R,R)-2 24
35h (S,S)-2
12
7
7
61/1
1/30
>99/1
1.3/1
1.2/1
PhCtC
24
(E)-2-Me-cinnamyl 35d (R,R)-2 24
10 (E)-crotyl
35e (S,S)-2
12
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC.
internal drb relative drc
entry catalyst time, h yield, %a
syn/anti
syn/anti
(entries 7 and 8). The additions to 2-methylcinnamalde-
hyde and to crotonaldehyde failed to provide the aldol
product under these conditions (entries 9 and 10).
1
2
3
4
5
6
7
(R,R)-1
(S,S)-1
(R,R)-39
40
2
2
8
7
7
5
7
87
80
65
76
82
67
79
>50/1
30/1
3/1
16/1
15/1
15/1
15/1
15/1
15/1
15/1
4. Aldol Additions of Ethyl Ketone Derived
Trichlorosilyl Enolates. 4.1. Aldol Addition of
Trichlorosilyl Enolate 23. The next stage of the
investigation involved the study of geometrically defined
ethyl ketone chiral enolates. With methods in hand to
prepare geometrically enriched enolates and enol ethers,
we began with a survey of catalysts using lactate derived
enolate (Z)-23 as the test substrate (Table 7). In addition
to catalyst (R,R)- and (S,S)-1, other structures surveyed
include the diazaphospholidines shown in Table 7. A
reaction concentration of 0.5 M was found to be optimal
allowing the aldol reaction to reach completion within 2
h using (R,R)-1 (Table 7, entry 1). Isolation of the crude
product followed by purification afforded (syn,syn)-41a
in 87% yield with greater than 50/1 internal diastereo-
selectivity for the (syn,syn)-adduct vs the (syn,anti)-
adduct (relative, internal). The relative diastereoselec-
tivity for the two new stereogenic centers formed during
the aldol reaction correlated appropriately with the
starting trichlorosilyl enolate geometry to give a 16/1 syn/
anti ratio.25
A survey of various chiral and achiral phosphoramides
showed a remarkable influence of the resident stereo-
genic center on the stereochemical course of the reaction
when using enolate (Z)-23 (Table 7, entries 3-7). Inter-
estingly, the configuration of the catalyst employed in
the aldol reaction had little effect on the stereochemical
outcome (entry 2). The use of (S,S)-1 led to only a minor
attenuation in internal diastereoselectivity from 50/1 to
30/1 for (R,R)-1 and (S,S)-1, respectively. Perhaps the
most profound demonstration of the directing influence
of the resident stereogenic center was seen with the
achiral N,N′-dimethylphospholidine catalysts, 29 and 40
(entries 4 and 5). In both cases, the internal diastereo-
selectivity was greater than 30/1. Remarkably, even the
use of hexamethylphosphoric triamide (HMPA) afforded
41a with an internal diastereoselectivity of 30/1 (entry
7). Among the catalysts employed only those bearing
34/1
37/1
3/1
29
30
HMPA
30/1
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC analysis. c Determined by 1H NMR analysis.
To establish the generality of the reaction of (Z)-23, a
survey of aldol additions to various aldehydes using
either (R,R)-1 or HMPA as catalysts was undertaken
(Table 8). To further optimize the reaction protocol, the
aldol additions were performed using in situ generated
trichlorosilyl enolate (Z)-23. This obviates the need for
handling of a sensitive compound, improving the overall
yield of aldol product with respect to the TMS enol ether.
Finally, to demonstrate the efficiency of the chiral
phosphoramides, aldol reactions were done using 5 mol
% of (R,R)-1, while still using 15 mol % of HMPA.
In the general in situ procedure, TMS enol ether
(Z)-14 was added dropwise to a solution of SiCl4 (2 equiv)
and 1 mol % Hg(OAc)2 in CH2Cl2, and the reaction was
monitored by 1H NMR spectroscopy. Following complete
conversion of the TMS enol ether, the contents of the
reaction vessel were placed under vacuum to remove the
volatile byproducts and solvent. A solution of phosphora-
mide in dry CH2Cl2 was then added via syringe. The reac-
tion vessel was cooled to -78 °C at which point freshly
distilled aldehyde was added dropwise via syringe.
Aromatic, heteroaromatic, olefinic and propargylic
aldehydes react under these conditions to provide the
(syn,syn)-aldol products in modest to good yields and high
diastereoselectivities. When HMPA was used as the
catalyst, the yields and selectivities were slightly attenu-
ated. The highest selectivity was observed with 1-naph-
thaldehyde affording the syn,syn-adduct exclusively.
To probe the generality of the Lewis base catalyzed
addition of 23 to aldehydes, addition of (E)-23 to benzal-
dehyde was studied to establish the difference, if any, in
rates between the chiral enolate geometrical isomers
(Scheme 12). Initial reactions at 5 mol % catalyst loadings
(25) The diastereoselectivities were assigned on the basis of chemical
correlation to the known methyl ester.
10830 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
TABLE 8. Aldol Addition of in Situ Generated
(Z)-24 to benzaldehyde were performed using several
different types of catalyst structures, and the results are
summarized in Table 9. The aldol addition catalyzed by
1 provided the syn (relative) aldol product in good yield
with good syn (internal) selectivity, although the E/Z
ratio of the enolate did not exactly reflect to the (relative)
diastereomeric ratio of the aldol product (entries 1 and
2). The survey of other catalysts also indicated the syn
(relative) diastereoselectivity was attenuated signifi-
cantly ranging from 4/1 to 11/1. Dimeric phosphoramide
42, which exhibited improved diastereoselectivities for
aldol addition of achiral ethyl ketone derived trichlorosi-
lyl enolates,11 showed the highest selectivity compared
to the monomer 1 (entry 6). The use of achiral catalysts
29 and 30 showed lower catalytic activity compared to
the chiral catalysts and also the relative diastereoselec-
tivity was decreased (entries 3 and 4). Phosphoramide
43 was also relatively unselective (entry 5). Dimeric
N-oxide 44, which was the most selective catalyst for the
addition of trichlorosilyl ketene acetal to ketones, showed
low catalytic activity and selectivity (entry 7).
Trichlorosilyl Enolate (Z)-23 to Aldehydes
loading
yield, drb (syn,syn)/
a
aldehyde R catalyst mol % product
%
minors
phenyl
(R,R)-1
HMPA
(R,R)-1
HMPA
5
15
5
15
5
15
5
15
5
15
5
15
5
15
41a
41a
41b
41b
41c
41c
41e
41e
41h
41h
41l
88
87
72
62
81
79
85
83
79
82
71
59
82
76
95/5
94/2/2/2
98/1/1
83/12/3/1
93/5/2
91/6/3
93/4/3
84/15/1
95/3/2
89/5/4/3
94/3/3
89/6/4/1
94/6
1-naphthyl
(E)-cinnamyl (R,R)-1
HMPA
(E)-crotyl
PhCtC-
2-naphthyl
2-furyl
(R,R)-1
HMPA
(R,R)-1
HMPA
(R,R)-1
HMPA
(R,R)-1
HMPA
TABLE 9. Aldol Addition of to Benzaldehyde Catalyzed
by Phosphoramides
41l
41n
41n
93/5/1
a Yield of chromatographically homogeneous material. b Deter-
mined by CSP-SFC analysis.
were atypically slow considering that benzaldehyde is
often a superlative acceptor. Furthermore, analysis of the
product revealed poor relative anti/syn stereoselectivity.
To improve the results for the (E)-enolate, the reactions
were repeated using a catalyst loading of 15 mol %.
Although a marginal improvement in reaction rates was
observed, the stereoselectivity remained modest relative
to that observed from (Z)-23.
loading, yield,
rel dr
int dr
a
entry
catalyst
mol %
%
(syn/anti)b (syn/anti)c
1
2
3
4
5
6
7
(R,R)-1
(S,S)-1
29
10
10
10
10
10
5
72
82
46
50
58
88
12
90/10
92/8
10/1
1/7.4
2.7/1
1.4/1
1/1.6
12/1
SCHEME 12
81/19
91/9
30
(R)-43
(R,R)-42
P-(R,R)-44
86/14
92/8
82/18
5
5.3/1
a Yield of chromatographically homogeneous material. b Deter-
mined by 1H NMR analysis. c Determined by CSP-SFC.
With the chiral catalysts (R,R)-1, (S,S)-1 and dimeric
phosphoramide (R,R)-42, the aldol additions of (Z)-24,
(Z)-25 and (E)-24 to benzaldehyde were studied (Table
10). Significantly higher relative and internal diastereo-
selectivities were achieved using TIPS-protected trichlo-
rosilyl enolate 25. The rate of the aldol addition for 25
was slower compared to the reactions with 24; however,
full conversion could be achieved within 8 h. The internal
diastereoselection for 25 was also determined using 29,
which indicated that a small 1,4-syn stereoinduction
results from the a-stereogenic center on the enolate. The
reversal of internal diastereoselection was again observed
by switching the catalyst configuration of 1. The matched
case was found to be the reaction with (R,R)-1 providing
the syn,syn-46a in excellent diastereoselectivity.
A brief survey of aldehyde acceptors was conducted
to assess the stereochemical outcome of the addition
of (E)-23. To allow for a direct comparison to the (Z)-
analogue, reactions were conducted using 5 mol % (R,R)-1
and allowed to proceed for 10-12 h. Unlike the broad
substrate generality illustrated with the (Z)-enolate, aldol
additions with (E)-23 afforded very modest selectivities.
Among the aldehydes surveyed, only benzaldehyde and
trans-cinnamaldehyde afforded the desired anti (relative)
products. Both 1-naphthaldehyde and 2-furaldehyde
provided the undesired syn (relative) isomer. In all cases,
internal diastereoselection was poor with the exception
of 1-naphthaldehyde which afforded a 6/1 mixture of
internal diastereomers.
4.2. Aldol Addition of Trichlorosilyl Enolates 24
and 25. Initial optimization studies for the addition of
On the other hand, the addition of the (E)-enolate was
not selective in either relative or internal sense. Despite
J. Org. Chem, Vol. 70, No. 26, 2005 10831

Denmark et al.
the high E/Z ratio of (E)-24, the reaction was only
slightly anti (relative) selective (Table 10, entries 4-6).
Also, the internal selectivities were poor, and the switch
in internal diastereoselection was not observed in these
cases. The use of dimeric catalyst 42 did not improve the
relative diastereoselectivity.
The absolute configuration of the aldol adduct 46c was
confirmed by comparing the physical data to the reported
synthetic intermediate in the literature.26 The configura-
tions of other aldol products were assigned by analogy
with the relative elution orders of diastereomers, using
CSP-SFC.
4.3. Aldol Addition of Trichlorosilyl Enolate 26
and 27. The optimization of reaction conditions for the
aldol addition of 26 and 27 was performed using benzal-
dehyde as the substrate (Table 12). A 10 mol % loading
of phosphoramide 1 efficiently catalyzed the addition to
provide the aldol products in moderate to good yields. The
aldol reaction of (Z)-enolates showed good syn (relative)
diastereoselectivities, indicating that the addition pro-
ceeds through chairlike transition structure. The benefi-
cial effect of the TIPS protecting group was again
observed in this case as it increased the diastereoselec-
tivity in both relative and internal senses. The addition
of (E)-27 was surprisingly syn (relative) selective. The
intrinsic selectivity of the chiral enolate (Z)-27 was again
determined using the achiral phosphoramide 29 (Table
12, entry 6).
To expand the scope of this aldol addition, (Z)-27 was
chosen to survey other aromatic and olefinic aldehydes,
and the results are summarized in Table 13. In all cases,
high syn (relative) diastereoselectivities were maintained.
Excellent syn (relative) diastereoselectivities were ob-
served for all the aldehydes surveyed. The internal
selectivity was variable depending on the aldehyde
structure. Exceptionally high diastereoselectivity was
observed for 48b compared to that of 48a. This trend may
be related to the bulk of the aldehyde substituent, and a
similar trend was observed for the addition of the ethyl
ketone enolate bearing an R-methyl stereogenic center.8c
A switch in the internal diastereoselection was ob-
served in all cases, depending on the catalyst configura-
tion.27 The internal selectivities were found to be signifi-
cantly higher for aromatic aldehydes compared to those
of other olefinic aldehydes (Table 13, entries 1 and 3).
Interestingly, the disparity in the internal selectivity
between (R,R)-1 (anti) and (S,S)-1 (syn) was larger in the
aromatic aldehydes than in the olefinic aldehydes. The
olefinic aldehydes showed modest internal diastereose-
lectivities ranging from 3/1 to 10/1 (Table 13, entries
5-10).
TABLE 10. Aldol Addition of (Z)-24, (E)-24, and (Z)-25 to
Benzaldehyde
yield,
rel dr
int dr
a
entry
enolate
catalyst
%
(syn/anti)b
(syn/anti)c
1
2
3
4
5
6
7
8
9
(Z)-24
(Z)-24
(Z)-24
(Z)-25
(Z)-25
(Z)-25
(E)-24
(E)-24
(E)-24
(R,R)-1
(S,S)-1
(R,R)-42
(R,R)-1
(S,S)-1
29
(R,R)-1
(S,S)-1
(R,R)-42
72
82
88
84
82
81
72
72
80
9/1
12/1
12/1
>19/1
>19/1
>19/1
1/4
10/1
1/7
12/1
24/1
1/8
5/1
6/1
1/2
2/1
1/2
9/1
a Yield of chromatographically homogeneous material. b Deter-
mined by 1H NMR analysis. c Determined by CSP-SFC.
Because the selectivities observed for the addition of
(Z)-25 to benzaldehyde were sufficiently high, this enolate
was chosen to examine the additions to other aromatic
and olefinic aldehydes (Table 11). Surprisingly, the
additions to these aldehydes under the previously estab-
lished conditions were too slow to achieve significant
conversion in a suitable time period. From previous
studies, it was found that an addition of various am-
monium salts can accelerate the aldol addition of trichlo-
rosilyl enolates, presumably by increasing the ionic
strength of the reaction medium. Therefore, 20 mol % of
tetrabutylammonium iodide (TBAI) was added to the
reaction mixture and the yields of the aldol products were
improved without diminishing the diastereoselectivities.
The syn (relative) selectivities were maintained in all
cases. The degree of relative diastereoselection varied
depending on aldehyde structure. By comparing a and
b, as well as e and g, the steric bulk of the aldehyde
substituent seems to have detrimental effect on the
relative diastereoselection (Table 11, entries 1-4 and
7-10). This phenomenon is contrary to the beneficial
effect of large aldehyde substituents in the addition of
methyl ketone enolates.8a A switch in the internal dias-
tereoselection was also observed in all cases, with (R,R)-1
affording the higher internal diastereomeric ratio. The
internal diastereoselectivities were found to be signifi-
cantly higher for aromatic aldehydes when compared to
olefinic aldehydes (Table 11, entries 1 and 3). The olefinic
aldehydes showed similar internal diastereoselectivities
ranging from 13/1 to 15/1 for the matched cases (Table
11, entries 5, 7, and 9). Additions to aliphatic aldehydes,
such as cyclohexanecarboxaldehyde and pivaldehyde
produced only trace amounts of aldol products under
these conditions. These aliphatic aldehydes form chloro-
hydrin species under these reaction conditions, and the
chlorohydrins are unreactive toward aldol additions with
trichlorosilyl enolates.7c
5. Aldol Additions of Ethyl Ketone Derived TMS
Enol Ethers. 5.1. Aldol Addition of Chiral TMS Enol
Ether 18. In the previous sections, the aldol additions
of trichlorosilyl enolates derived from ethyl ketones were
shown to selectively provide syn (relative) diastereomers;
however, the corresponding anti diastereomers could not
be obtained selectively. By analogy to the anti-selective
addition of propanoate derived silyl ketene acetals,14 it
should be possible to develop an anti (relative) selective
aldol addition of R-substituted trialkylsilyl enol ethers
(26) Feutrill, J. T.; Lilly, M. J.; Rizzacasa, M. A. Org. Lett. 2002, 4,
525-528.
(27) The absolute configuration of the aldol product syn,anti-48a was
determined by chemical correlation. First, the aldol product was
converted to the corresponding ester 49 by Baeyer-Villiger oxidation
with buffered m-CPBA (Bernhard, W.; Fleming, I. J. Organomet. Chem.
1984, 271, 281-288). Next, 49 was hydrolyzed under basic conditions
to provide diol 50 in good yield. The configuration was assigned as
(1S,2R)-1-phenylpropane-1,2-diol by comparing the sign of the optical
rotation to the literature value. (a) Kihumbum, D.; Stillger, T.;
Hummel, W.; Liese, A. Tetrahedron: Asymmetry 2002, 13, 1069-1072.
10832 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
TABLE 11. Aldol Addition of (Z)-25 to Various Aldehydesa
entry
aldehyde R
product
catalyst
time, h
yield, %b
rel dr (syn/anti)c
int dr (syn/anti)d
1e
2e
3
phenyl
46a
46a
46b
46b
46c
46c
46e
46e
46g
46g
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
8
8
84
82
71
79
88
75
90
85
85
80
>19/1
>19/1
14/1
24/1
1/8
89/1
1/17
14/1
1/6
15/1
1/5
13/1
1/5
phenyl
1-naphthyl
1-naphthyl
(E)-cinnamyl
(E)-cinnamyl
(E)-crotyl
(E)-crotyl
MeCHdCMe
MeCHdCMe
8
4
8
14/1
5
10
10
6
9/1
15/1
6
7
>19/1
>19/1
13/1
8
6
9
7
10
7
19/1
1
a (Z)-25 contained ∼10% of bisenoxysilane. b Yield of chromatographically homogeneous material. c Determined by H NMR analysis.
d Determined by CSP-SFC. e No TBAI was added.
TABLE 12. Aldol Additions of 26 and 27 to
Benzaldehyde
iodide indeed increased the yield. The effect was clear in
the reaction catalyzed by monomeric catalyst 1 wherein
the yield increased dramatically (entry 5). For the reac-
tion catalyzed by dimeric phosphoramide 2, the effect of
the ammonium salt was relatively small; however, only
a 10% increase in yield was observed without eroding the
diastereoselectivity (entries 6 and 7).
To increase the yield of the aldol addition, the reaction
temperature for the aldol addition of 18 was varied (Table
15). An increase in yield was observed as the reaction
temperature was raised to -50 °C; however, the reaction
at -20 °C failed to afford the aldol product. Interestingly,
as the temperature of the reaction increased, the relative
diastereoselectivity dropped significantly, whereas the
internal diastereoselectivity remained almost constant.
Reaction at -60 °C was selected as the optimal reaction
temperature because the diastereoselectivity was only
slightly attenuated. The yield of the aldol product was
further improved by using an excess (1.5 equiv) of the
aldehyde. Because the enol ether now has become the
limiting reagent, diisopropylethylamine was added as an
acid scavenger to prevent the destruction of the nucleo-
phile by adventitious acid that may occur during the
reaction. Under the optimal conditions, up to an 81%
yield of aldol product can be obtained.
The scope of aldehyde in this reaction was surveyed
next. However, under the same conditions described
above, the reaction of (Z)-18 with cinnamaldehyde did
not provide the addition product and the starting materi-
als were recovered. Thus, the reaction conditions seem
to be only applicable to the addition to benzaldehyde.
5.2. Aldol Addition of (S)-3-Hydroxybutyrate De-
rived TMS Enol Ethers 21 and 22. The aldol addition
of TMS enol ether derived from an ethyl ketone bearing
a â-silyloxy stereogenic center was also investigated using
bisphosphoramide catalysis. Using the similar conditions
established in the previous section, the aldol addition of
(Z)-21 to benzaldehyde was executed (Scheme 13). The
aldol product was obtained in good yield under catalysis
by (R,R)-2. However, it was surprising to find that the
aldol addition provided the syn (relative) diastereomer
as the major product. By the use of the monomeric
phosphoramide 1, the reaction was moderately anti-
yield,
rel dr
int dr
a
entry enolate Z/E catalyst
%
(syn/anti)b (syn/anti)c
1
2
3
4
5
6
(Z)-26
(Z)-26
(Z)-27
(Z)-27
(E)-27
12/1 (R,R)-1
12/1 (S,S)-1
16/1 (R,R)-1
16/1 (S,S)-1
1/15 (S,S)-1
59
60
84
86
80
83
6/1
12/1
1/14
14/1
1/16
10/1
1/1
>19/1
>19/1
3/1
(Z)-27 >19/1 29
>19/1
1/1.4
a Yield of chromatographically homogeneous material. b Deter-
mined by 1H NMR analysis. c Determined by CSP-SFC.
derived from ketones using the bisphosphoramide/SiCl4
system. Therefore, the TMS enol ethers derived from the
chiral ethyl ketones have been assayed in the bisphos-
phoramide-catalyzed Mukaiyama-type aldol additions.
The initial reaction optimization surveyed catalyst
structure and the effect of tetraalkylammonium salts for
the aldol addition of (Z)-18 to benzaldehyde (Table 14).
With 5 mol % of the dimeric catalyst 2, the aldol product
was obtained in modest yields although the reaction did
not go to completion even after 24 h. The monomeric
phosphoramide 1 was not catalytically active under these
conditions (entries 3 and 4). When (S,S)-2 was used, the
internal diastereoselectivity was only 6/1 within the anti
(relative) manifold (entry 2). With (R,R)-2 as a catalyst,
the anti,anti diastereomer was formed almost exclusively
(entry 1).28
In aldol additions of the corresponding trichlorosilyl
enolate, the addition of tetraalkylammonium salts was
found to increase the yields of the reactions. Therefore,
the effect of additives has been investigated in this aldol
reaction. Addition of 20 mol % of tetrabutylammonium
(28) The assignments for the relative anti diastereomers were made
by analogy to the assignment of aldol products from ref 7c.
J. Org. Chem, Vol. 70, No. 26, 2005 10833

Denmark et al.
TABLE 13. Aldol Addition of (Z)-27 to Various Aldehydesa
entry
aldehyde R
product
catalyst
time, h
yield, %b
rel dr (syn/anti)c
int dr (syn/anti)d
1e
2e
3
phenyl
48a
48a
48b
48b
48c
48c
48d
48d
48e
48e
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-1
6
6
84
86
80
75
83
81
83
81
79
83
>19/1
>19/1
>19/1
>19/1
>19/1
>19/1
>19/1
>19/1
>19/1
>19/1
1/16
10/1
1/30
10/1
1/10
8/1
phenyl
1-naphthyl
1-naphthyl
8
4
8
5
(E)-cinnamyl
(E)-cinnamyl
(E)-2-Me-cinnamyl
(E)-2-Me-cinnamyl
(E)-crotyl
12
12
8
6
7
1/3
8
8
3/1
9
8
1/7
10
(E)-crotyl
8
6/1
a (Z)-27 contained 10% of the bisenoxysilane, and the Z/E ratio was >19/1. b Yield of chromatographically homogeneous material.
c Determined by 1H NMR analysis. d Determined by CSP-SFC. e Z/E ratio was 16/1.
TABLE 14. Aldol Addition of (Z)-18 Catalyzed by Chiral
TABLE 15. Aldol Addition of 18 at Higher Reaction
Temperaturesa
Phosphoramidesa
temp,
°C
yield,
rel dr
int dr
yield,
rel dr
int dr
b
entry
catalyst
%
(syn/anti)c
(syn/anti)d
c
entry catalyst additiveb
%
(syn/anti)d (syn/anti)e
1
(R,R)-2
(R,R)-2
(R,R)-2
(R,R)-2
(S,S)-2
-78
-60
-50
-60
-60
55
62
75
81
72
1/>19
1/>19
1/17
1/>19
1/>19
1/34
1/37
1/35
1/74
5.6/1
1
2
3
4
5
6
7f
(R,R)-2
(S,S)-2
(R,R)-1
(S,S)-1
(R,R)-1
(S,S)-2
(R,R)-2
40
46
tr
1/>19
1/>19
1/>50
2e
3
6/1
4f
5f
tr
TBAI
TBAI
TBAI
55
50
55
1/>19
1/>19
1/>19
1/34
8/1
1/>50
a Z/E ratio of 18 was >19/1. b Yield of chromatographically
homogeneous material. c Determined by 1H NMR analysis. d De-
termined by CSP-SFC. e 5 mol % of 2 was used. f Excess benzal-
dehyde (1.5 equiv) was used and 20 mol % of i-Pr2EtN was added.
a Z/E ratio for 18 was >19/1. b 20 mol % added. c Yield of
chromatographically homogeneous material. d Determined by 1H
NMR analysis. e Determined by CSP-SFC. f TBS-protected 17 was
used as nucleophile.
SCHEME 13
selective, though the yield of the aldol product was
attenuated. Within the anti (relative) manifold, the
internal diastereoselectivity was only modest.
Although the enolate geometry did not affect the
diastereoselectivity of aldol addition in propanoate de-
rived ketene acetals, the aldol addition of TIPS-protected
(E)-22 was attempted to examine the diastereoselectivity.
Interestingly, the aldol addition catalyzed by (S,S)-2
showed good relative anti selectivity (17/1); however, the
yield of the aldol product was dramatically decreased.
This observation is perplexing since there was no reactiv-
ity difference in the aldol addition of (Z)- and (E)-silyl
ketene acetal derived from propanoate.7c
The aldol additions of chiral TMS enol ethers (Z)-21
and (E)-22 to benzaldehyde gave disappointing results.
The addition of (Z)-21 was not selective under these
reaction conditions and provided a complex mixture of
all four possible isomers. The addition of (E)-22 was
relative anti-selective but did not afford synthetically
useful yields.
Discussion
1. Aldol Addition of Methyl Ketone Derived
Trichlorosilyl Enolates 9 and 12. The aldol additions
of 9 and 12 are efficiently catalyzed by phosphoramide
1. Unlike the aldol addition of the lactate derived
trichlorosilyl enolate,9 the catalyst controlled diastereo-
selectivity was observed for both of these enolates. For
9, there is a weak inherent (substrate-controlled) stere-
oinduction from the resident stereogenic center (entry 12,
Table 2). This stereoinduction may be explained by the
following two models (Figure 1). Transition structure I
places the least sterically demanding group (hydrogen)
facing toward the silicon to avoid the steric crowding at
10834 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
the silicon center. This model predicts the approach of
benzaldehyde from the face containing the methyl group
rather than that containing the oxygenated substituent.
The other model, II, proposes the coordination of the
trichlorosilyl ether by the OTBS group. Although the TBS
group is often considered as nonchelating protecting
group, the nature of electrophilic silicon in this species
is not well-known, and this proposal may be appropriate.
Thus, the approach of the aldehyde is controlled by the
difference between a hydrogen or a methyl group whereby
approach from the hydrogen side is consistent with the
observed selectivity.
process is mechanistically distinct from the addition of
trichlorosilyl enolates, and thus, the nature and extent
of diastereoselection will be different. Indeed, the aldol
addition of TMS enol ether 8 showed significantly higher
diastereoselectivity than the aldol addition of the corre-
sponding trichlorosilyl enolate 9 for benzaldehyde (Table
5). Interestingly, the anti diastereomer was obtained in
the matched case using (S,S)-2. The internal diastereo-
selectivity was not measured directly using achiral
phosphoramide; however, it is clear from the results that
the stereoinduction from the resident stereogenic center
is small. Thus, the catalyst-controlled diastereoselection
dominated allowing for selective preparation of both
diastereomers.
The stereoinduction of the â-stereogenic center is also
absent in the addition of TMS enol ether 11 (Scheme 11).
Unlike the corresponding to the addition of trichlorosilyl
enolate 12, the aldol adducts of benzaldehyde were
obtained in excellent diastereoselectivities.
The observed facial selectivity of the aldehyde is
consistent with the aldol additions of achiral nucleophiles
using the same catalyst system. The calculated model
predicts the complex shown in Figure 2 is one of the
lowest energy species.14d In this model, (R,R)-2 blocks the
Si face of the aldehyde, and the nucleophile can prefer-
entially attack from the Re face.
FIGURE 1. Transition state model for the aldol addition of
9.
The R-stereogenic center exhibited a modest, intrinsic
1,4-syn stereoinduction, but the 1,4-syn selectivity could
be increased dramatically by catalysis with (R,R)-1. The
external stereoinduction from the catalyst dominated the
stereochemical course of the aldol addition, allowing for
selective preparation of either diastereomer by simply
changing the catalyst configuration. The reaction was
found to be general for various aldehydes including
aromatic, olefinic, acetylenic and aliphatic aldehydes
which are traditionally problematic substrates under
these reaction conditions.
The aldol addition of 12 showed that the stereoinduc-
tion from â-stereogenic center is almost nonexistent
(Scheme 10). This results contrasts to the recent reports
on the aldol additions of boron enolates bearing a
â-stereogenic center.12 In these cases, a strong 1,5-anti
stereoinduction is observed. However, these studies also
revealed that the nature of the protecting group at the
â-hydroxyl group significantly affects the substrate-
controlled stereoinduction. The changing the substituent
at the â-position from alkyl ethers (benzyl and 4-meth-
oxybenzyl) to TBS ether lead to almost complete loss of
the diastereoselectivity. Therefore, it is conceivable that,
in the aldol addition of 12, the substrate-controlled
diastereoselectivity may be increased by changing to, for
example, a benzyl ether.
The absence of internal stereoinduction allowed the
catalyst-controlled diastereoselection for the aldol addi-
tions of 12. Using chiral phosphoramide 1, either dias-
tereomer of 35a could be obtained; however, the diaste-
reoselectivities are unacceptably low. By changing the
protecting group on â-hydroxyl from TBS to TIPS, the
diastereoselectivity was decreased. Because the observed
selectivities for benzaldehyde were not high, the scope
of the aldehyde for the aldol addition of 12 was not
investigated. However, these types of aldol products are
obtained in high yields and selectivities using the TMS
enol ether 11 as the nucleophile (Table 6).
FIGURE 2. SiCl4-(R,R)-2-PhCHO complex.
The aldehyde scope for the addition of TMS enol ethers
is relatively limited compared to the corresponding aldol
additions of trichlorosilyl enolates. The aromatic and
cinnamaldehyde provided aldol products in high yields
and excellent selectivities. However, the additions to
simple olefinic aldehydes such as crotonaldehyde pro-
vided a complex mixture of products, and the desired
aldol adducts can only be obtained in low yields. Interest-
ingly, the presence of R-branching in the aldehyde led to
significant decrease in the reactivity and selectivity.
These results contrasts to the aldol additions of trichlo-
rosilyl enolates where these aldehydes showed increased
selectivities. Presumably, the added steric bulk around
the carbonyl hinders the coordination of the aldehyde to
the catalyst complex in these aldehydes.
For aliphatic aldehydes, the chlorosilyl ether formation
inhibits the aldol addition.14a When stronger nucleophiles
such as ketene acetals and isocyanides are used, the
addition to aliphatic aldehydes was observed under
similar conditions. Lack of reactivity for silyl enol ethers
toward aliphatic aldehydes confirms the less nucleophilic
nature of trialkylsilyl enol ethers.
2. Aldol Additions of Methyl Ketone Derived TMS
Enol Ethers. The aldol addition of trialkylsilyl enol
ethers is catalyzed by the Lewis acid generated by
activating SiCl4 with various chiral phosphoramides. This
J. Org. Chem, Vol. 70, No. 26, 2005 10835

Denmark et al.
3. Aldol Additions of Ethyl Ketone Derived
Trichlorosilyl Enolates. Earlier studies on the aldol
additions of R-substituted trichlorosilyl enolates showed
that the enolate geometry is strictly reflected in the
relative diastereoselectivity, which is consistent with a
Zimmerman-Traxler-type transition state model.29 There-
fore, the geometrically enriched (Z)-enolate should pro-
vide the syn (relative) diastereomer exclusively, whereas
the cyclic enolates ((E)-enolates) provide the anti (rela-
tive) diastereomer as the major product when chiral
phosphoramide 1 is used.
model, the aldehyde now approaches from the re face of
the enolate, leading to the syn,anti diastereomer.
SCHEME 14
The aldol addition of lactate derived enolate (Z)-23
under phosphoramide catalysis provides the syn,syn-aldol
product in high selectivity (Table 8). A survey of chiral
and achiral phosphoramide demonstrated the near per-
fect syn/anti-Z/E correlation indicating that the chairlike
transition structure is maintained. Even the bulky phos-
phoramide 39, which catalyzes the aldol addition pref-
erentially through a boat transition structure, provided
the syn product.
The relative diastereoselectivity for the addition of
(Z)-24 did not strictly mirror the Z/E ratio of the enolate
(Table 10). This result indicates that there are other
competitive transition structures (possibly boat-shaped
models) leading to anti (relative) diastereomers and
resulting in erosion of relative diastereoselectivity. How-
ever, the relative diastereoselectivity is improved by
changing the protecting group on â-hydroxyl group from
TBS to TIPS group. With (Z)-25, the syn (relative)
diastereomers are obtained exclusively.
The internal diastereoselection varied significantly
with catalyst structure (Table 10). The 1,2-stilbenedi-
amine derived catalyst 1 provided the best internal
diastereoselection. Catalyst (R,R)-1 showed 1,4-syn (in-
ternal) stereoinduction yielding the syn,syn diastereomer
as the major product, whereas the catalyst (S,S)-1
provided the syn,anti diastereomer selectively. The in-
ternal diastereoselection from the chiral enolate was
determined using achiral catalyst 29, and showed a very
weak 1,4-syn stereoinduction. Thus, the reaction with
(R,R)-1 represents the matched case affording the syn,syn
diastereomer with high selectivity. The external diaste-
reoselection from the catalyst dominates in this reaction
allowing for catalyst controlled preparation of either syn
(relative) diastereomer.
As was the case for methyl ketone enolate 12, the effect
of the resident stereogenic center in the aldol addition
of (Z)-27 was negligible, inducing nearly 1/1 internal
diastereoselection (Table 12). This outcome allowed di-
astereocontrolled aldol addition to produce either syn
(relative) diastereomer by changing the configuration of
1. The use of (R,R)-1 produced the syn,anti diastereomer,
while the syn,syn diastereomer was obtained using the
(S,S)-1. Thus, the external stereoinduction from the
catalyst determined the diastereoselection in these aldol
additions. This catalyst-controlled aldol addition en-
hances the synthetic utility of trichlorosilyl enolates.
For the aldol additions of (Z)-trichlorosilyl enolates, the
scope in the aldehyde was proven to be broad, and
aromatic and conjugated aldehydes afforded the syn
The internal stereoselectivities vary for different cata-
lyst structure. For the stilbene-1,2-diamine derived cata-
lyst, (R,R)-1, an excellent internal syn selectivity can be
obtained. The use of the enantiomeric catalyst (S,S)-1 is
unable to reverse the sense of internal selectivity, and
the aldol product is obtained with good internal syn
selectivity. These observations indicate the strong influ-
ence of the resident stereogenic center, and the stereo-
chemical course of the aldol addition is determined solely
by this stereocenter. To support this explanation, the ad-
ditions using various achiral phosphoramides including
HMPA demonstrate that the internal syn aldol product
is preferentially formed under catalyzed conditions regard-
less of the catalyst configuration. The bulky phosphora-
mides such as (R,R)-39 show attenuated internal dias-
tereoselectivity when compared to other phosphoramides.
The observed stereochemical outcome can be explained
by the nonchelation model that places the OTBS sub-
stituent on the enolate in plane with the enolate double
bond to minimize the dipole moment (Scheme 14).16 The
two enolate faces are differentiated by the size of the
groups on the stereogenic center (H vs Me), and the
aldehyde approaches from the less hindered Si face of
the enolate. The chairlike arrangement of the aldehyde
leads to the formation of the observed syn,syn diastere-
omer. The low selectivities observed for the bulky phos-
phoramides can be rationalized by the other transition
structure II.8b These phosphoramides are known to favor
boatlike transition structures via a mechanism that
involves only one phosphoramide in the stereodetermin-
ing step.30 In this pentacoordinate species, it is possible
that the silyloxy group could coordinate the Lewis acidic
silicon to form an octahedral, cationic silicon intermedi-
ate. This internal coordination may favor the chairlike
arrangement over the usual boatlike transition structure
for these phosphoramides. Although the coordinating
ability of the TBS ether is modest at best,31 the proximity
of the silyloxy group to the cationic silicon is believed to
enhance the possibility for this type of chelation.32 In this
(31) For a discussion of the coordinating abilities of various ether
substituents, see: (a) Reetz, M. T. Acc. Chem. Res. 1993, 26, 462-
468. (b) Chen, X. N.; Hortellano, E. R.; Eliel, E. L.; Frye, S. V. J. Am.
Chem. Soc. 1990, 112, 6130-6131. (c) Mori, S.; Nakamura, M.;
Nakamura, E.; Koga, N.; Morokuma, K. J. Am. Chem. Soc. 1995, 117,
5055-5056.
(32) For a discussion of chelation as a stereocontrol element in
lactate derived stannous enolate, see: Paterson, I.; Tillyer, R. D.
Tetrahedron Lett. 1992, 33, 4233-4236.
(29) Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957,
79, 1920-1923.
(30) (a) Denmark, S. E.; Su, X.; Nishigaichi, Y. J. Am. Chem. Soc.
1998, 120, 12990-12991. (b) Denmark, S. E.; Pham, S. M. Helv. Chim.
Acta 2000, 83, 1846-1853.
10836 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
(relative) products selectively using the chiral phosphora-
mide 1. The aldol addition of (Z)-23 to various aldehydes
revealed that the strong effect of the resident stereogenic
center dominated the stereochemical outcome, and the
reactions catalyzed by HMPA exhibited high selectivity
for the syn,syn diastereomers. For (Z)-25 and (Z)-27, the
weak or almost negligible internal stereoinduction al-
lowed the catalyst control to dominate the overall stere-
oinduction, which in turn provided access to both of the
syn (relative) diastereomers selectively for various alde-
hydes. However, these nucleophiles failed to react with
aliphatic aldehydes. This is presumably because the
added steric bulk of R-substituent on the enolate de-
creases the reactivity.
nantly anti (Table 15). Unlike the addition of the
analogous trichlorosilyl enolates, the formation of anti
(relative) product from the (Z)-enol ether indicates the
involvement of an open transition structure. This obser-
vation is consistent with the previous studies that
employed propanoate derived silyl ketene acetals, wherein
the geometry of the ketene acetal did not influence the
relative diastereoselectivity of the aldol addition.14d
The internal diastereoselectivity differed dramatically
between the two enantiomeric catalysts (Table 15). Thus,
the reaction catalyzed by (R,R)-2 is apparently the
matched case providing anti,anti diastereomer in high
stereoselectivity, and it is clear that the internal stereo-
induction from the R-stereogenic center showed strong
1,4-syn stereoinduction. However, the diastereoselection
was switched between these two enantiomeric catalysts,
indicating that the primary diastereoselection is provided
by the external stereoinduction from the catalyst.
In contrast, the aldol addition of TMS enol ether
(Z)-21 was unselective. Only modest diastereoselectivities
were observed for the reactions with chiral phosphora-
mides 1 and 2. The relative diastereoselectivities could
be improved by changing the double bond geometry of
the enol ether; however, this change resulted in loss of
reactivity. These results are unexpected considering the
additions of the propanoate derived ketene acetals are
insensitive to the double bond geometry both in terms of
the diastereoselectivity and the reactivity.
Overall, the aldol addition of chiral TMS enol ethers
18 and 21 under bisphosphoramide catalysis provided the
access to the anti (relative) diastereomers. The reaction
was slow compared to the analogous TMS enol ether
derived from methyl ketones, and the increased catalyst
loading and temperature were required to achieve sig-
nificant conversion. This is presumably due to the low
reactivity of the TMS enol ether. For this process to be
synthetically useful, the development of a more active
catalyst and/or the use of a more reactive nucleophile
need to be investigated.
5. Summary. These aldol additions using the different
classes of chiral trichlorosilyl enolates and TMS enol
ethers represent interesting cases of double diastereo-
differentiating aldol additions. Given the number of
permutations investigated in this study, a summary of
the key findings is warranted:
On the other hand, reactions with the E-configured
trichlorosilyl enolate demonstrated only modest stereo-
selectivity even with chiral phosphoramides. The lack of
stereoinduction by either the resident stereogenic center
or the chiral phosphoramide appears to be a common
problem with (E)-trichlorosilyl enolates.1k Although it
may typically be expected that the reaction of (E)-enolates
would be more facile due to the orientation of the enolate
substituent being placed in a pseudoequatorial position
as observed in the aldol addition of cyclic trichlorosilyl
enolates,33 the results from the reactions of (E)-trichlo-
rosilyl enolates derived from acyclic ketones would sug-
gest otherwise. This appears counterintuitive since re-
actions of an (E)-enolate would obviate the axial orienta-
tion of the enolate substituent. Based upon experimental
results, the position of the enolate substituent in an (E)-
enolate has negative repercussions on the overall ste-
reoselectivity of the aldol addition. Unlike typical con-
formational analyses of cyclohexanes, the transition
structures for the aldol additions contain no severe 1,3-
diaxial interactions for a (Z)-enolate substituent posi-
tioned axially in a chairlike arrangement. Conversely,
an equatorial substituent from an (E)-enolate would
provide 1,2-gauche interactions. To minimize these un-
favorable steric interactions, the aldehyde substituent
can be oriented axially to afford syn aldol products in the
relative manifold. Since the axial orientation of the
aldehyde substituent affords 1,3-diaxial interactions with
the enolate spectator group, the overall effect is that both
transition structures become similar in energy, resulting
in the loss of stereoselectivity.34 In addition, positioning
of the enolate substituent in an E-conformation allows
the metal center to rotate away from the bulky spectator
substituents, minimizing steric encumbrances. This now
allows the components to adopt alternative transition
structures, altering the predicted stereochemical course
of the reaction.35 The poor stereoselectivities observed for
the reactions of the (E)-enolates to aldehydes certainly
suggest multiple transition structures are operative.
(1) In the aldol addition of methyl ketone derived
trichlorosilyl enolates 9 and 12, the resident stereogenic
centers exhibited modest and weak stereoinduction re-
spectively, leading to dominant catalyst controlled selec-
tivity. The scope of aldehyde structure was examined
with 9, and various aldehydes including aromatic, ole-
finic, acetylenic, and aliphatic aldehydes provided aldol
products in modest to good diastereoselectivities.
(2) The additions of methyl ketone derived TMS enol
ethers 8 and 11 catalyzed by bisphosphoramide 2 showed
extremely high diastereoselectivities. In this family, the
aldehyde scope is limited to aromatic aldehydes. In all
cases, catalyst-controlled diastereoselection was achieved
allowing for access to either the 1,5-syn or -anti diaste-
reomer.
4. Aldol Additions of Ethyl Ketone Derived TMS
Enol Ethers. In the aldol addition of (Z)-18 to benzal-
dehyde catalyzed by bisphosphoramide 2, the relative
configuration of the major aldol product was predomi-
(33) Stavenger, R. A., Ph.D. Thesis, University of Illinois at Urbana-
Champaign, Urbana, Illinois, 1999.
(3) In the aldol additions of ethyl ketone derived
enolates 23, 25 and 27, the addition of the (Z)-enolates
afforded predominantly the syn (relative) diastereomer.
On the other hand, aldol addition of the corresponding
(34) Heathcock, C. H. In Comprehensive Carbanion Chemistry;
Buncel, E., Durst, T., Eds.; Elsevier: Amsterdam, 1984; Part B,
Chapter 4.
(35) Bernardi, A.; Comotti, A.; Gennari, C.; Hewkin, C. T.; Goodman,
J. M.; Schlapbach, A.; Paterson, I. Tetrahedron 1994, 50, 1227-1242.
J. Org. Chem, Vol. 70, No. 26, 2005 10837

Denmark et al.
28a). Trichlorosilyl enolate 9 (140 mg, 0.40 mmol, 1.2 equiv)
was added to a solution of (R,R)-1 (12.2 mg, 0.033 mmol, 0.10
equiv) in 0.66 mL of CH2Cl2. The reaction mixture was cooled
using an acetone/CO2 bath, and PhCHO (34 µL, 0.33 mmol, 1
equiv) was added using a syringe. The reaction was monitored
by taking an aliquot from the reaction mixture and analyzing
on TLC (hexane/EtOAc, 5/1). After 1 h of stirring (or when
the aldehyde spot disappeared), the reaction mixture was
quenched by pouring into a vigorously stirring 5 mL of cold
saturated NaHCO3 aq. The resulted slurry was stirred for 15
min and filtered through Celite. The layers were separated,
and the aqueous layer was extracted with 10 mL of CH2Cl2.
The organic phases were combined, washed with 2 mL of brine
and dried over sodium sulfate. After filtration, the filtrate was
concentrated. The crude product was purified by column
chromatography (hexane/EtOAc, 5/1), and 78 mg (0.24 mmol,
73%) of 28a was obtained as a clear viscous oil. 1H NMR: (500
MHz, CDCl3) syn-28a 7.37-7.26 (m, 5 H, aromatic); 5.17 (dt,
J ) 8.5, 3.2, 1 H, HC(5)); 3.75-3.65 (ABX, 2 H, H2C(1)); 3.50
(d, J ) 2.9, 1 H, HO); 2.98-2.84 (ABX, 2 H, H2C(4)); 2.78 (sext,
J ) 7.1, 1 H, HC(2)); 1.03 (d, J ) 7.1, 3 H, H3C(6)); 0.87 (s,
9H, C(9)); 0.04 (d, J ) 1.7, 6 H, C(7)); anti-28a 7.37-7.26 (m,
5 H, aromatic); 5.14 (dt, J ) 9.3, 3.0, 1 H, HC(5)); 3.77-3.65
(ABX, 2 H, H2C(1)); 3.54 (d, J ) 3.0, 1 H, HO); 2.98-2.84 (ABX,
2 H, H2C(4)); 2.79 (sext, J ) 7.1, 1 H, HC(2)); 1.04 (d, J ) 7.1,
3 H, H3C(6)); 0.88 (s, 9H, C(9)); 0.04 (d, J ) 1.7, 6 H). SFC:
(OD column, 175 bar, 2.5 mL/min, 1.7% MeOH) tR syn-28a
5.414 min (91.3%); tR anti-28a 5.989 min (8.7%).
(E)-enolates was not selective. The stereochemical course
in the addition of (Z)-23 was primarily controlled by the
strong induction from the resident stereogenic center.
The syn,syn diastereomer was obtained in high selectivity
regardless of the catalyst configuration and structure. On
the other hand, the aldol additions of (Z)-25 and (Z)-27
showed catalyst-controlled diastereoselection. With these
enolates, aromatic and olefinic aldehydes react to provide
the syn (relative) diastereomers in good diastereoselec-
tivities.
The aldol additions of TMS enol ethers (Z)-18 and
(Z)-21 were anti (relative) selective. However, the lack
of reactivity under the SiCl4/bisphosphoramide catalysis
was apparent, and the search for a more active catalyst
system and better reaction conditions is needed.
Conclusion
The aldol additions of chiral silyl enol ethers were
investigated under Lewis base catalysis. The aldol ad-
ditions of chiral trichlorosilyl enolates were efficiently
catalyzed by chiral phosphoramide 1, whereas the addi-
tions of chiral TMS enol ethers were catalyzed by dimeric
phosphoramide 2. In aldol additions of lactate derived
trichlorosilyl enolates, the internal stereoinduction from
R-silyloxy group dominated the overall diastereoselection.
On the other hand, the effects of R-methyl and â-silyloxy
groups were relatively small, allowing for catalyst-
controlled diastereoselection. A wide variety of aldehydes
including aromatic and alkenyl aldehydes reacted under
phosphoramide catalysis, providing stereodyads and
triads that resemble common structural motifs found in
natural products. Further studies are currently under-
way to develop reagents capable of reacting with aliphatic
aldehydes and to apply these methods in the synthesis
of complex natural products.
General Procedure III. Aldol Addition of in Situ
Generated Trichlorosilyl Enolate Using 8: (1S,4R)-5-tert-
Butyldimethylsililoxy-4-methyl-1-hydroxy-1-phenyl-3-
pentanone (syn-28a). To a suspension of mercury(II) acetate
(3.2 mg, 0.01 mmol, 0.01 equiv) in 1 mL of methylene chloride
was added silicon tetrachloride (0.23 mL, 2.0 mmol, 2.0 equiv)
and the trimethylsilyl enol ether 8 (288 mg, 1.00 mmol). The
reaction mixture was stirred at room temperature for 30 min.
The excess silicon tetrachloride and the solvent were removed
under vacuum. To the residue was added solution of the chiral
phosphoramide (R,R)-1 (36.9 mg, 0.10 mmol, 0.10 equiv) in 2
mL of methylene chloride via a cannula. The reaction mixture
was cooled to -78 °C before the addition of benzaldehyde (102
µL, 1.00 mmol, 1 equiv). The reaction mixture was stirred at
-78 °C for 2.5 h and then was quenched by pouring into 5 mL
of vigorously stirring cold saturated aqueous sodium bicarbon-
ate. The resulting slurry was stirred for 1 h and was filtered
through Celite. The layers were separated and the aqueous
layer was extracted with 20 mL of methylene chloride. The
combined organic extracts were washed with 5 mL of brine,
dried over sodium sulfate and concentrated. The crude product
was chromatographed (hexane/ether, 5/1, SiO2) to give 28a
Experimental Section
General Experimental. See Supporting Information.
Experimental Procedures. General Procedure I. MeLi-
Mediated Transsilylation of 17: (2Z,4S)- 5-(tert-Butyl-
dimethyl-silyloxy)-4-methyl-3-trichlorosilyloxy-2-pen-
tene ((Z)-24). In a 25-mL two-neck flask was placed a solution
of (Z)-17 (908 mg, 3.00 mmol) in ether (6 mL) at 0 °C. To the
solution was slowly added MeLi (3.00 mL, 4.50 mmol, 1.5
equiv, 1.5 M in ether). The reaction mixture was stirred for
4.5 h at room temperature, then cooled to -70 °C using an
dry ice/2-propanol bath. The reaction mixture was transferred
dropwise using a cannula to a 100-mL two-neck flask contain-
ing a cold (-70 °C) solution of silicon tetrachloride (3.45 mL,
30.0 mmol, 10 equiv) in ether (6 mL). The reaction mixture
was stirred at -70 °C for 1 h and gradually warmed to room
temperature. The precipitate settled at the bottom of the flask,
and the supernatant was transferred to a 35-mL round-bottom
flask. The volatiles were removed under vacuum (0.5 mmHg),
and the residue was distilled using a Kugelrohr apparatus (150
°C ABT at 1.0 mmHg), to afford 901 mg (2.48 mmol, 83%) of
(Z)-24 as a clear colorless oil. 1H NMR: (CDCl3, 500 MHz) 4.53
(qd, J ) 6.8, 0.5, 1 H, HC(2)); 3.69-3.43 (ABX, 2 H, H2C(5));
2.40 (sext, J ) 6.3, 1 H, HC(4)); 1.60 (dd, J ) 7.1, 0.7, 3 H,
H3C(1)); 1.07 (d, J ) 7.1, 3 H, H3C(6)); 0.88 (s, 9 H, H3C(9));
0.03 (s, 6 H, H3C(7)) 13C NMR: (CDCl3, 126 MHz) 150.8 (C(3));
106.4 (C(2)); 65.3 (C(5)); 42.1 (C(4)); 25.8 (C(9)); 18.2 (C(8));
14.9 (C(7)); 11.0 (C(1)); -5.5, -5.4 (C(7)).
1
(268 mg, 0.83 mmol, 83%) as viscous colorless oil. H NMR:
(500 MHz, CDCl3) 7.23-7.47 (m, 5H, aromatic); 5.17 (dt, J )
8.8, 3.2, 1H, HC(1)); 3.65-3.76 (ABX, 2H, C(5)H2); 3.49 (d, J
) 2.9, 1H, OH); 2.86-2.96 (ABX, 2H, C(2)H2); 2.78 (m,1H, C(4)-
H); 1.03 (d, J ) 6.9, 3H, C(6)H3); 0.87 (s, 9H, (C(9)H3)3); 0.04
(d, J ) 2, 6H, C(7)H3). 13C NMR: (125 MHz, CDCl3) 214.7
(C(3));142.9, 128.5, 127.5, 125.6 (aromatic); 69.7 (C(1)); 65.6
(C(5)); 51.4 (C(2)); 49.0 (C(4)); 25.8 (C(9)); 18.2 (C(8)); 12.6
(C(6)); -5.6 (C(7)); IR: (neat) 3467 (m, br); 3064 (w); 3031 (w);
2954 (s); 2931 (s); 2885 (m); 2858 (m); 1708 (m); 1463 (m); 1388
(m); 1256 (m); 1100 (s); 838 (s). SFC: (Chiralpak OD column,
150 bar, 3.0 mL/min, 1.4% MeOH) tR (2S,5R)-syn-28a 4.74 min
(95%); (2S,5S)-anti-28a 5.21 min (5%). Anal. Calcd for C18H30O3-
Si (322.51) C 67.03; H 9.38. Found: C 66.94; H 9.42.
General Procedure IV. Aldol Addition of in Situ
Generated Trichlorosilyl Enolate Using 11: (1R,5S)-5-
tert-Butyldimethylsililoxy-1-hydroxy-1-phenyl-3-hex-
anone (anti-35a). To a suspension of mercury(II) acetate (3.2
mg, 0.01 mmol, 0.01 equiv) in 1 mL of methylene chloride was
added silicon tetrachloride (0.23 mL, 2.0 mmol, 2.0 equiv) and
the trimethylsilyl enol ether 11 (289 mg, 1.00 mmol). The
reaction mixture was stirred at room temperature for 30 min.
General Procedure II. Aldol Reaction of Isolated
Trichlorosilyl Enolate: (2S,5R)-1-tert-Butyldimethylsi-
liloxy-2-methyl-5-hydroxy-5-phenyl-3-pentanone (syn-
10838 J. Org. Chem., Vol. 70, No. 26, 2005

Chiral Lewis Base Catalyzed Aldol Additions
The excess silicon tetrachloride and the solvent were removed
under vacuum. To the residue was added a solution of the
chiral phosphoramide (R,R)-1 (36.9 mg, 0.10 mmol, 0.10 equiv)
in 2 mL of methylene chloride via a cannula. The reaction
mixture was cooled to -78 °C before the addition of benzal-
dehyde (102 µL, 1.00 mmol, 1 equiv). The reaction mixture
was stirred at -78 °C for 4 h and then was quenched by
pouring into 5 mL of vigorously stirring cold saturated aqueous
sodium bicarbonate. The resulting slurry was stirred for 3 h
and was filtered through Celite. The layers were separated
and the aqueous layer was extracted with 20 mL of methylene
chloride. The combined organic extracts were washed with 5
mL of brine, dried over sodium sulfate and concentrated. The
crude product was chromatographed (pentane/ether, 3/1, SiO2)
to give 35a (198 mg, 0.61 mmol, 61%) as viscous colorless oil.
1H NMR: (500 MHz, CDCl3) 7.35-7.34 (m, 4 H, HC(8), HC-
(9)); 7.29-7.26 (m, 1 H, HC(10)); 5.15 (dt, J ) 8.6, 3.5, 1H,
HC(1)); 4.32 (sext, J ) 6.1, 1 H, HC(5)); 3.36 (d, J ) 3.2, 1 H,
OH); 2.92-2.82 (ABX, 2H, H2C(2)); 2.67-2.41 (ABX, 2H, H2C-
(4)); 1.17 (d, J ) 6.2, 3H, H3C(6)); 0.86 (s, 9H, H3C(13)); 0.05
(d, J ) 12.4, 6H, H3C(11)). 13C NMR: (125 MHz, CDCl3) 210.6
(C(3));142.7 (C(7)); 128.5 (C(9)); 127.6 (C(10)); 125.6 (C(8)); 69.6
(C(5)); 65.6 (C(1)); 53.3 (C(4)); 52.9 (C(2)); 25.7 (C(13)); 24.0
(C(6)); 17.9 (C(12)); -4.5, -5.0 (C(11)). IR: (neat) 3438 (m, br);
3064 (w); 3032 (w); 2956 (s); 2929 (s); 2895 (m); 2850 (m); 1711
(s); 1495 (w); 1462 (m); 1377 (m); 1255 (m); 1134 (m); 1090
(m); 1066 (s); 1041 (m); 1007 (m). SFC: tR ) 4.695 min (OD,
150 bar, 3 mL/min, 2% MeOH). Anal. Calcd for C18H30O3Si
(322.51): C, 67.03; H, 9.38. Found: C, 66.94; H, 9.52%
to give a cloudy oil. A solution of (R,R)-1 (18 mg, 0.05 mmol,
0.05 equiv) in CH2Cl2 (2.0 mL) was then added via cannula
and the mixture was cooled to -78 °C. Benzaldehyde (102 µL,
1.0 mmol) was then added dropwise via syringe and the
reaction mixture was allowed to stir at -78 °C for 10 h. The
reaction mixture was then poured into a rapidly stirring
saturated aqueous NaHCO3 solution (30 mL) submerged in
an ice bath and was allowed to stir at room temperature for 6
h. The heterogeneous mixture was then filtered through Celite,
the organic phase was separated, and the aqueous phase was
extracted with CH2Cl2 (3 × 15 mL). The organic extracts were
combined, dried over Na2SO4, filtered and concentrated to give
a crude oil. Purification by column chromatography (SiO2,
pentane/Et2O, 6/1) afforded 284 mg (88%) of 41a as a clear,
colorless oil. The diastereomeric ratio was determined to be
1
(syn,syn)-41a/minor isomers, 95/5 by SFC analysis. H NMR:
(CDCl3, 500 MHz) 7.38-7.20 (m, 5 H, 2 × H(C2′′), 2 × HC-
(3′′), HC(4′′)); 5.06 (dd, J ) 5.0, 2.8, 1 H, HC(5), syn,syn); 5.01
(dd, J ) 5.0, 2.8, 1 H, HC(5), anti,syn); 4.77 (dd, J ) 8.5, 4.2,
1 H, HC(5), anti-relative); 4.73 (dd, J ) 8.5, 4.2, 1 H, HC(5),
anti-relative); 4.19 (q, J ) 6.9, 1 H, HC(2)); 3.37 (dq, J ) 7.2,
5.0, 1 H, HC(4)); 3.25, (d, J ) 2.8, 1 H, OH); 1.27 (d, J ) 6.9,
3 H, H3C(1)); 1.05 (d, J ) 7.2, 3 H, H3C(1′)); 0.90 (s, 9 H, H3C-
(3′′′)); 0.08 (s, 3 H, H3C(1′′′)); 0.06 (s, 3 H, H3C(1′′′)). 13C NMR:
(CDCl3, 125 MHz) 218.7 (C(3)); 141.7 (C(1′′)); 128.2 (C(3′′));
127.2 (C(4′′)); 126.0 (C(2′′)); 75.7 (C(2)); 72.8 (C(5)); 46.9 (C(4));
25.7 (C(3′′′)); 21.1 (C(1)); 18.0 (C(8)); 10.4 (C(1′)); -4.7 (C(1′′′));
-5.0 (C(1′′′)). SFC: tR (2S,4R,5S)-41a, 5.1 min (Daicel Chiral-
pak AD, 5% MeOH in CO2, 150 bar, 40 °C, 3.0 mL min-1).
General Procedure V. Aldol Addition of Methyl Ke-
tone Derived TMS Enol Ethers: (1R,4S)-5-tert-Butyldim-
ethyl-sililoxy-4-methyl-1-hydroxy-1-phenyl-3-pen-
tanone (syn-28a). In a 10-mL Schlenk flask was placed
(R,R)-2 (42 mg, 0.05 mmol, 0.05 equiv) in dichloromethane (1
mL). To the solution was added benzaldehyde (0.102 mL, 1.00
mmol), and the solution was cooled to -78 °C (bath temper-
ature) using a dry ice/acetone bath. To the solution was added
silicon tetrachloride (0.172 mL, 1.50 mmol, 1.5 equiv), and the
solution was stirred for 5 min. To the solution was added
dropwise 8 (346 mg, 1.20 mmol, 1.2 equiv), and the reaction
mixture was stirred at -78 °C (bath temperature) for 4 h. The
reaction mixture was quenched by pouring into a vigorously
stirring cold (0 °C) saturated aqueous NaHCO3 solution (15
mL), and the resulting emulsion was stirred for at least 2 h.
The emulsion was filtered through a pad of Celite, and the
filtrate was transferred into a 60-mL separatory funnel. The
biphasic mixture was extracted with dichloromethane (2 × 20
mL), and the combined extracts were washed with brine (10
mL). The solution was dried over Na2SO4 and concentrated
under reduced pressure. The resulting crude oil was chro-
matographed (silica gel, pentane/ether, 3/1, 30 mm) to afford
294 mg (0.91 mmol, 91%) of 28a as a clear colorless oil. 1H
NMR: (500 MHz, CDCl3) 7.23-7.47 (m, 5H, aromatic); 5.17
(dt, J ) 8.8, 3.2, 1H, HC(1)); 3.65-3.76 (ABX, 2H, C(5)H2);
3.49 (d, J ) 2.9, 1H, OH); 2.86-2.96 (ABX, 2H, C(2)H2); 2.78
(m,1H, C(4)H); 1.03 (d, J ) 6.9, 3H, C(6)H3); 0.87 (s, 9H, (C(9)-
H3)3); 0.04 (d, J ) 2, 6H, C(7)H3). 13C NMR: (125 MHz, CDCl3)
214.7 (C(3));142.9, 128.5, 127.5, 125.6 (aromatic); 69.7 (C(1));
65.6 (C(5)); 51.4 (C(2)); 49.0 (C(4)); 25.8 (C(9)); 18.2 (C(8)); 12.6
(C(6)); -5.6 (C(7)). SFC: (Chiralpak OD column, 125 bar, 3.0
mL/min, 2.0% MeOH) tR (1R,4S)-syn-28a 4.797 min (95.9%);
(1S,4S)-anti-28a 5.271 min (4.1%).
General Procedure VII. Aldol Addition of (Z)-24:
(1R,2R,4S)-5-tert-Butyldimethylsilyloxy-2,4-dimethyl-1-
hydroxy-1-phenyl-3-pentanone (syn,syn-45a). In a 10-mL
Schlenk flask was placed a solution of (R,R)-1 (37 mg, 0.10
mmol, 0.10 equiv) in CH2Cl2 (1 mL). To the solution was added
(Z)-24 (364 mg, 1.00 mmol) and the solution was cooled to -70
°C using an acetone/CO2 bath. To the solution was added
dropwise PhCHO (102 µL, 1.00 mmol, 1 equiv) and the reaction
mixture was stirred for 6 h at -70 °C. The reaction mixture
was quenched by pouring into a vigorously stirring cold
saturated aqueous NaHCO3 solution (5 mL). The resulting
slurry was stirred for 1 h and filtered through Celite. The
layers were separated, and the aqueous layer was extracted
with CH2Cl2 (2 × 10 mL). The combined organic extracts were
washed with brine (5 mL), dried over sodium sulfate and
concentrated under reduced pressure. The crude product was
purified by column chromatography (silica gel, pentane/ether,
3/1, 30 mm) to afford 241 mg (0.72 mmol, 72%) of 45a as a
clear viscous oil. 1H NMR: (500 MHz, CDCl3) 7.34-7.32 (m, 4
H, HC(7), HC(8)); 7.25 (m, 1 H, HC(9)); 5.22 (t, J ) 2.4, 1 H,
HC(1)); 3.75-3.60 (ABX, 2 H, H2C(5)); 3.45 (d, J ) 2.4, 1 H,
HO); 3.02 (sext, J ) 5.4, 1 H, HC(4)); 2.97 (qd, J ) 7.1, 2.7, 1
H, HC(2)); 1.02 (d, J ) 6.8, 3 H, H3C(11)); 1.01 (d, J ) 7.3, 3
H, H3C(10)); 0.89 (s, 9H, H3C(14)); 0.06 (d, J ) 3.2, 6 H, H3C-
(12)). 13C NMR: (126 MHz, CDCl3) 219.2 (C(3)); 141.7 (C(6));
128.1 (C(8)); 127.1 (C(9)); 125.8 (C(7)); 71.9 (C(5)); 66.3 (C(1));
52.8 (C(2)); 47.3 (C(4)); 25.9 (C(14)); 18.3 (C(13)); 13.2 (C(11));
8.8 (C(10)); -5.6 (C(12)). IR: (neat) 3469 (br, w); 2954 (m);
2931 (m); 2858 (m); 1702 (m); 1462 (m); 1389 (w); 1255 (m);
1097 (m). SFC: tR ) 1.834 min (AD, 150 bar, 3 mL/min, 5%
MeOH) Anal. Calcd for C19H32O3Si (336.54): C, 67.81; H, 9.58.
Found: C, 67.74; H, 9.69.
General Procedure VIII. Aldol Additions of Trichlo-
rosilyl enolate (Z)-26: (1R,2R,5S)-5-tert-Butyldimethyl-
silyloxy-2-methyl-1-hydroxy-1-phenyl-3-hexanone (syn,an-
ti-47a). In a 10-mL Schlenk flask was placed a solution of
(R,R)-1 (37 mg, 0.10 mmol, 0.10 equiv) in CH2Cl2 (1 mL). To
the solution was added (Z)-26 (364 mg, 1.00 mmol) and the
solution was cooled to -68 °C using i-PrOH/CO2 bath. To the
solution was added PhCHO (102 µL, 1.00 mmol, 1 equiv) and
the reaction mixture was stirred for 6 h at -68 °C. The reaction
mixture was quenched by pouring into a vigorously stirring
cold saturated aqueous NaHCO3 solution (15 mL). The result-
General Procedure VI: Aldol Addition of in Situ
Generated Lactate Derived Enolate (Z)-23: (+)-(2S,4R,5S)-
5-Hydroxy-4-methyl-5-phenyl-2-[((dimethyl)-(1,1-dimeth-
ylethyl)-silyl)oxy]-3-pentanone (41a). Silyl enol ether 14
(273 mg, 1.0 mmol) was added quickly to a stirred suspension
of silicon tetrachloride (230 µL, 2.0 mmol, 2.0 equiv) and
mercuric acetate (3.2 mg, 0.01 mmol, 0.01 equiv) in CH2Cl2
(1.0 mL) at room temperature. After addition, the mixture was
stirred at room temperature for 18 h, and then the volatile
components were removed under reduced pressure (0.1 mmHg)
J. Org. Chem, Vol. 70, No. 26, 2005 10839

Denmark et al.
ing slurry was stirred for 3 h and filtered through Celite. The
layers were separated, and the aqueous layer was extracted
with CH2Cl2 (2 × 15 mL). The combined extracts were washed
with brine (5 mL), dried over sodium sulfate and concentrated.
The crude oil was purified by column chromatography (silica
gel, pentane/ether, 3/1, 30 mm) to afford 200 mg (0.59 mmol,
59%) of 47a as a clear viscous oil. 1H NMR: (500 MHz, CDCl3)
7.36-7.30 (m, 4 H, HC(8), HC(9)); 7.25 (m, 1 H, HC(10)); 5.15
(t, J ) 2.9, 1 H, HC(1)); 4.32 (sext, J ) 6.6, 1 H, HC(5)); 3.16
(d, J ) 2.6, 1 H, HO); 2.81 (qd, J ) 7.3, 3.1, 1 H, HC(2)); 2.72-
2.43 (ABX, 2 H, H2C(4)); 1.16 (d, J ) 6.1, 3 H, H3C(6)); 1.04
(d, J ) 7.3, 3 H, H3C(11)); 0.87 (s, 9H, H3C(14)); 0.06 (d, J )
11.2, 6 H, H3C(12)). 13C NMR: (126 MHz, CDCl3) 214.7 (C(3));
141.6 (C(7)); 128.2 (C(9)); 127.2 (C(10)); 125.8 (C(8)); 72.3 (C(5));
65.4 (C(1)); 52.5 (C(2)); 51.3 (C(4)); 25.8 (C(14)); 23.9 (C(6));
17.9 (C(13)); 9.2 (C(11)); -4.6, -4.9 (C(12)). IR: (neat) 3458
(br, m); 3064 (w); 3032 (w); 2956 (m); 2931 (m); 2895 (m); 2858
(m); 1707 (m); 1454 (m); 1375 (m); 1255 (m); 1134 (m); 1095
(m); 1005 (m). SFC: tR ) 2.441 min (OD, 150 bar, 3 mL/min,
5% MeOH). Anal. Calcd for C19H32O3Si (336.54): C, 67.81; H,
9.58. Found: C, 67.61; H, 9.59.
General Procedure IX. Aldol Addition of TMS Enol
Ether 25: (1R,2S,4S)-1-Hydroxy-2,4-dimethyl-1-phenyl-
5-triisopropylsilanyloxy-pentan-3-one (anti,anti-46a). In
a 10-mL Schlenk flask were placed (R,R)-2 (84 mg, 0.10 mmol,
0.10 equiv), tetrabutylammonium iodide (74 mg, 0.20 mmol,
0.2 equiv) in dichloromethane (2 mL). To the solution were
added benzaldehyde (0.152 mL, 1.50 mmol, 1.5 equiv) and
diisopropylethylamine (0.035 mL, 0.20 mmol, 0.2 equiv), and
the solution was cooled to -72 °C (internal temperature) using
a dry ice/2-propanol bath. To the solution was added silicon
tetrachloride (0.172 mL, 1.50 mmol, 1.5 equiv), and the
solution was stirred for 5 min. To the solution was added
dropwise 25 (345 mg, 1.00 mmol), and the reaction mixture
was stirred at -60 °C (internal temperature) for 24 h. The
reaction mixture was quenched by pouring into a vigorously
stirring cold (0 °C) saturated aqueous NaHCO3 solution (15
mL), and the resulting emulsion was stirred for at least 3 h.
The emulsion was filtered through a pad of Celite, and the
filtrate was transferred into a 60-mL separatory funnel. The
biphasic mixture was extracted with dichloromethane (2 × 20
mL), and the combined extracts were washed with brine (10
mL). The solution was dried over Na2SO4 and concentrated
under reduced pressure. The resulting crude oil was chro-
matographed (silica gel, pentane/ether, 4/1, 30 mm) to afford
307 mg (0.81 mmol, 81%) of 46a as a clear colorless oil. 1H
NMR: (500 MHz, CDCl3) 7.36-7.26 (m, 5 H, 2 × HC(7), 2 ×
HC(8), HC(9)); 4.78 (dd, J ) 8.0, 4.8, 1 H, HC(1)); 3.78 (ABX,
2 H, 2 × HC(5)); 3.09 (d, J ) 4.6, 1 H, HO); 3.05 (pent, J )
7.6, 1 H, HC(2)); 2.89 (sextd, J ) 5.1, 1.2, 1 H, HC(4)); 1.04
(m, 21 H, 3 × HC(12), 18 × HC(13)); 0.97 (d, J ) 7.1, 3 H, 3
× HC(10)); 0.93 (d, J ) 7.1, 3 H, 3 × HC(11)). 13C NMR: (126
MHz, CDCl3) 217.9 (C(3)); 142.3 (C(6)); 128.3 (2 × C(7)); 127.7
(C(9)); 126.5 (2 × C(8)); 76.5 (C(1)); 65.5 (C(5)); 53.1 (C(2));
49.1 (C(4)); 17.88, 17.89 (6 × C(13)); 13.8 (C(10)); 12.8 (C(11));
11.8 (3 × C(12)). IR: (neat) 3452 (br, m); 3064 (w); 3032 (w);
2943 (s); 2893 (m); 2868 (s); 1711 (m); 1495 (w); 1462 (m); 1383
(m); 1248 (m); 1101 (m); 1066 (m); 1011 (m). SFC: (Chiralcel-
OD, 125 bar, 3 mL/min, 5.0% MeOH) (anti,anti)-46a, tR
)
4.194 min (96.5%); (anti,syn)-46a, tR ) 3.311 min (1.3%);
(syn,syn)-46a, tR ) 3.712 min (1.4%); (syn,anti)-46a, tR ) 4.979
min (0.7%). Anal. Calcd for C22H38O3Si (378.62): C, 69.79; H,
10.12. Found: C, 69.86; H, 10.31.
Acknowledgment. We are grateful for generous
financial support from the National Science Foundation
(NSF CHE0105205 and CHE 0414440). S.F. thanks
Abbott Laboratories and the Johnson and Johnson
Pharmaceutical Research Institute for graduate fellow-
ships. S.M.P thanks the Eastman Kodak Company for
a graduate fellowship.
Supporting Information Available: Experimental de-
tails for the preparation and full spectroscopic and analytical
characterization of all compounds reported, including CIF files.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO051930+
10840 J. Org. Chem., Vol. 70, No. 26, 2005