Direct catalytic asymmetric aldol reactions promoted by novel heterobimetallic catalysts possessing strong Br?nsted base: A new strategy for the development of Lewis acid-Br?nsted base bifunctional catalysts

Article abstract of DOI:10.1016/S0040-4020(01)00070-9

Novel heterobimetallic catalysts possessing lithium alkoxide as Br?nsted base have been developed. The catalysts were found to be effective in the direct asymmetric aldol reaction of less acidic dialkyl ketones as well as aromatic ketones, affording the aldol products in moderate yield and ee. Attempts at diastereo- and enantioselective direct catalytic aldol reaction of ethyl ketone using small molecule catalysts are reported for the first time.

Full text of DOI:10.1016/S0040-4020(01)00070-9

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 2569±2579
Direct catalytic asymmetric aldol reactions promoted by novel
heterobimetallic catalysts possessing strong Brùnsted base:
a new strategy for the development of Lewis acid±Brùnsted
base bifunctional catalysts
Naoki Yoshikawa and Masakatsu Shibasakip
Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
This paper is dedicated to Professor Henri B. Kagan on the occasion of him being awarded the 1999 Tetrahedron Prize
Received 1 August 2000; revised 4 October 2000; accepted 9 October 2000
AbstractÐNovel heterobimetallic catalysts possessing lithium alkoxide as Brùnsted base have been developed. The catalysts were found to
be effective in the direct asymmetric aldol reaction of less acidic dialkyl ketones as well as aromatic ketones, affording the aldol products in
moderate yield and ee. Attempts at diastereo- and enantioselective direct catalytic aldol reaction of ethyl ketone using small molecule
catalysts are reported for the ®rst time. q 2001 Published by Elsevier Science Ltd.
1. Introduction
cient catalysts for this method have been reported until
recently,2±5 presumably due to dif®culties described
below. For example, the catalyst must not only activate
the aldehyde in order to get high ee but also has to abstract
an a-hydrogen of the ketone to generate an enolate. Retro-
aldol reactions and self-condensation of aldehydes must be
prevented as well. Having established the system for Lewis
acid±Brùnsted base bifunctional catalysts,6 we have
achieved direct catalytic asymmetric aldol reactions using
LaLi3tris(binaphthoxide) (LLB),7 and subsequently using a
barium complex (BaB±M)8 and a heteropolymetallic cata-
lyst.9 Another group has recently reported a proline-cata-
lyzed reaction.10 These reactions have excluded the
The development of a catalytic asymmetric aldol reaction is
one of recent landmarks in organic chemistry. A number of
excellent catalysts for asymmetric syntheses of aldol
adducts from latent enolates such as enol silyl ethers or
enol methyl ethers have been reported (Scheme 1).1 In
these reactions, however, latent enolates must be prepared
prior to the reaction, and stoichiometric amounts of bases
and reagents such as (CH3)3SiCl are necessary. Therefore,
the development of a direct catalytic asymmetric aldol reac-
tion, in which the pre-activation of nucleophiles is no longer
needed, has been worthy of notice. Nevertheless, no ef®-
Scheme 1.
Keywords: direct catalytic asymmetric aldol reaction; lanthanide; heterobimetallic multifunctional catalyst.
Corresponding author. Tel.: 181-3-5684-0651; fax: 181-3-5684-5206; e-mail: mshibasa@mol.f.u-tokyo.ac.jp
p
0040±4020/01/$ - see front matter q 2001 Published by Elsevier Science Ltd.
PII: S0040-4020(01)00070-9

2570
N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
2. Results and discussions
In 1999, we developed a heteropolymetallic catalyst by
addition of potassium hydroxide (KOH) and H2O to
LLB.9 The catalyst was found to accelerate the direct asym-
metric aldol reaction of ketones with aldehydes. Fig. 1
shows the postulated structure of the heteropolymetallic
catalyst. Potassium hydroxide was thought to coordinate
to LLB and to act as Brùnsted base for the abstraction of
a-hydrogens of ketones. Since the rate-determining step
proved to be the deprotonation of ketones on the basis of
kinetic studies, the modi®cation of Brùnsted base seemed to
be necessary to enhance the catalytic activity. Scheme 2
shows the possible design of new catalysts, which consist
of a center metal (M) acting as a Lewis acid and metal-
alkoxide moiety (OM0) acting as a strong Brùnsted base.
The novel strategy for designing catalysts is to attach the
Brùnsted base moiety covalently to the naphthyl ring
through a carbon tether,11 different from the cases of
conventional heterobimetallic catalysts6 and the heteropoly-
metallic catalyst.9 We believe this strategy facilitates the
optimization of bifunctional catalysts, since the Brùnsted
base moiety can be varied without changing the framework
of the complexes.
Figure 1. Structure of the heteropolymetallic catalyst.
2.1. Direct catalytic asymmetric aldol reactions of
methyl ketones
Our ®rst aim was to synthesize catalysts which possess two
lithium alkoxide moieties, from ligand 6 (Scheme 3 shows
an assumed structure).12 Scheme 4 summarizes the sequence
for the synthesis of ligand 6 starting from MOM-protected
(R)-BINOL. Two hydroxyethyl groups were introduced at
3,30-position of 1,10-binaphthyl by treatment of lithiated
MOM-BINOL with an excess amount of ethylene oxide in
the presence of BF3´OEt2.13 After one of two hydroxyl func-
tions was protected as allyl ether,14 the MOM group was
replaced with a benzyl group. The remaining hydroxyl
group was oxidized to aldehyde using Dess±Martin
periodinane to afford 5. The formation of a symmetrical
ether was achieved by reductive coupling of the aldehyde
using triethylsilane in the presence of trimethylsilyl
tri¯ate.15 The deprotection of the allyl group was performed
Scheme 2.
necessity of pre-activation of ketones for catalytic asym-
metic aldol reactions. Modest to excellent results were
obtained in the reactions between aldehydes and methyl
ketones such as acetophenone or acetone. On the other
hand, there is no report that these catalysts are capable of
promoting the aldol reactions of esters or methylene ketones
such as 3-pentanone. Thus we were prompted to develop
novel catalysts to broaden the generality of the direct cata-
lytic asymmetric aldol reaction. In this article, we report a
novel strategy, which allows a ¯exible design of new cata-
lysts. This strategy is applied to the development of direct
asymmetric aldol reactions.
Scheme 3.

N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
2571
Scheme 4. Synthetic route to ligand 6.
Table 1. Effect of center metals and solvents
Entry
Metal (M) source
BuLi (equiv. to M)
Solvent
Temp. (8C)
Time (h)
Yield (%)
ee (%)
1
2
3
4
5
6
7
8
Ti(O-i-Pr)4
Me3Al
2
3
2
3
3
3
3
3
THF
THF
THF
THF
CH3CN
CH2Cl2
Toluene
Toluene
2208C
2208C
Rt
2208C
2208C
2208C
2208C
2208C
28
28
48
0.75
21
21
0.75
2
Trace
31
16
75
58
Trace
61
68
±
13
23
6
10
±
Zr(O-i-Pr)4
La(O-i-Pr)3
La(O-i-Pr)3
La(O-i-Pr)3
La(O-i-Pr)3
Yb(O-i-Pr)3
22
±1
by palladium-catalyzed reduction using tributyltin hydride
and zinc chloride16 to afford the alcohol. Ligand 6 was
obtained after cleavage of the benzyl ethers under Birch's
conditions in the absence of a proton source.
depends on the choice of center metal of the catalyst (M
in Scheme 3). Although titanium (Ti) complex had been
expected to show high catalytic activity, based on previous
examples of asymmetric Mukaiyama±aldol reactions, the
direct aldol reaction between aldehyde 7 and acetophenone
(8a) was not promoted by Ti±Li complex (entry 1). Neither
catalyst containing Al nor Zr as center metal was applicable
(entries 2, 3). In contrast, moderate to good yields were
achieved when the catalyst was prepared from 6, La(O-i-
Pr)3 and BuLi. Toluene proved to be the best solvent, giving
the product in 61% yield with 22% ee (entry 7). The reaction
was completed in 45 min at 2208C. Using Yb, the complex
gave unsatisfactory result (21% ee) (entry 8).
With the desired ligand in hand, we prepared several kinds
of catalysts by treatment of ligand 6 with 1 mol equiv. of
metal isopropoxide (or Me3Al) and 2±3 mol equiv. of
n-BuLi. Table 1 indicates that the reactivity strongly
Table 2. Effect of the amount of base
With the best center metal, we turned our attention to the
effect of the Brùnsted base part of the catalyst. The impor-
tance of lithium alkoxide moiety bonded to naphthalene ring
has been clearly demonstrated by varying the equivalent of
BuLi used for the catalyst preparation. Table 2 shows the
results of the aldol reaction promoted by La±Li catalysts in
toluene. When only 2 mol equiv. of BuLi were used (the
Entry
BuLi (x mol equiv. to La)
Time (h)
Yield (%)
ee (%)
1
2
3
3 mol equiv.
2 mol equiv.
1 mol equiv.
0.75
19
19
61
51
0
22
4
±

2572
N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
Table 3. Effect of lithium salts
entry 5). The addition of LiI gave rise to a slightly higher ee
(entry 6).
²,³
We were encouraged by these results and attempted to tune
the ligand by changing the length of the tether or the linker
(m, n in Scheme 3). In fact, the length proved to affect both
reactivity and selectivity signi®cantly (Table 4). The use of
a shorter tether of the alcohol (ligand 10) was found to
decrease the reactivity (entry 2). Moreover, a catalyst with
a shorter linker (ligand 11) showed much lower selectivity
as well as lower reactivity (entry 3). Consequently, catalyst
A proved to be the best catalyst.
Entry
Li-salit Temperature (8C) Time (h) Yield (%) ee (%)
1
2
3
4
±
2208C
2208C
2208C
2208C
2308C
2308C
0.75
4
61
30
61
84
70
70
22
4
LiCl
LiBr
LiI
±
1
1.5
4
36
34
60
67
5a (cat A)
6a (cat A) LiI
4
With the best catalyst in hand, we tried to reduce the catalyst
loading (Table 5). The reactions were examined using
2 mol equiv. of the ketone. As a result, we were glad to
®nd that as little as 3 mol% of catalyst A promoted the
reaction with comparable yield and ee (entry 4). In the
presence of less than 3 mol% of catalyst, the reaction did
not proceed well. The relatively low reactivity would be due
to the decomposition of the catalyst since the lithium
alkoxide part could be easily hydrolyzed by a small amount
of water generated from dehydration of the product.
a
BuLi solution provided by Aldrich was used for the catalyst preparation.
Table 4. Aldol reactions using several types of ligands
Compared to aromatic ketones such as acetophenone, it is
more dif®cult to carry out the direct aldol reaction of ali-
phatic ketones such as 2-butanone or pinacolone, due to
their lower acidity. To demonstrate the activity of catalyst
A, several aliphatic ketones were tested. Table 6 shows that
this catalyst can promote the reaction of less acidic aliphatic
ketones as well as aromatic ketones (entries 2, 3). It is
noteworthy that the aldol reaction of isopropyl methyl
ketone (8c) gave the product (9c) in moderate yield (entry
3), since this substrate has not been reported as a good
substrate for direct aldol reactions. It is well known that
a,b-unsaturated ketones can form vinylogous enolates and
undergo condensations at the g-position under basic condi-
tions. Catalyst A was found to be rather effective in the
reaction of a,b-unsaturated ketone 8d with aldehyde 7.
The ketone reacted smoothly and solely at the a-position
to afford the corresponding product (9d) in 66% yield and
60% ee (entry 4).
Entry
Ligand
Temperature (8C) Time (h) Yield (%) ee (%)
1
2
3
6 (cat A) 2308C
10
11
4
19
22
70
55
48
67
35
3
2208C
2208C
BuLi solution provided by Aldrich was used for the catalyst preparation.
best catalyst is prepared from 3 mol equiv. of BuLi), the
reaction time was prolonged to 19 h (cf. 0.75 h for the
control experiment (entry 1)) and the enantiomeric excess
decreased to 4% (entry 2). Moreover, a catalyst which lacks
aliphatic alkoxide (entry 3) gave no product at all.
Next, the effect of lithium salts17 as additive was investi-
gated (Table 3). When lithium chloride (LiCl) (1 mol equiv.
to the catalyst) was added, the reactivity and the enantio-
selectivity dropped (entry 2). On the other hand, lithium
bromide (LiBr) and lithium iodide (LiI) were found to
improve the enantioselectivity slightly without signi®cant
drop of reactivity (entries 3, 4). When the catalyst was
prepared from BuLi provided by Aldrich, which contains
only trace amounts of lithium salts, to exclude LiCl, much
higher enantiomeric excess (60% ee) was achieved (cat A,
On the basis of the observed effects of the center metal
(Table 1) and the lithium alkoxide moiety (Table 2), the
reaction seems to be promoted by the cooperation of the
Lewis acid and the Brùnsted base, similar to the conven-
tional catalyses.7±9 We describe below a postulated
mechanism for the catalysis (Scheme 5). A molecular
model of catalyst A suggests that an intramolecular coordi-
nation of the linker oxygen12a and a lithium alkoxide moiety
to the center metal can take place, thus stabilizing the mono-
meric structure (I). The remaining alkoxide moiety abstracts
a proton from the a-position of the ketone to give an alcohol
and the corresponding lithium enolate. A chelation of Li to
the alcohol and to an oxygen atom of BINOL would ®x the
lithium enolate inside the cavity of the catalyst to form
Table 5. Effect of catalyst amount
Entry
Catalyst (mol %)
Time (h)
Yield (%)
ee (%)
1
2
3
4
20
10
5
4
19
28
28
70
76
61
66
60
58
74
63
²
The moderate yields are due to unreacted aldehyde and a small amount of
dehydrated product.
Several enolizable aldehydes (e.g. isobutyraldehyde) instead of 7 were
also tested. However, they suffered from poor enantioselectivities and
lower chemical yields.
³
3
BuLi solution provided by Aldrich was used for the catalyst preparation.

N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
Table 6. Aldol reactions of several kinds of ketones
2573
Entry
Ketone
Temp. (8C)
Product
9a (RˆPh)
Time (h)
Yield (%)
ee (%)
1
2
8a (2 mol equiv.)
8b (5 mol equiv.)
2308C
2208C
4
70
71
67
40
9b (RˆEt)
14
3
8c (5 mol equiv.)
8d (5 mol equiv.)
2208C
9c (Rˆi-Pr)
71
62
45
4
5
2308C
2208C
9d (RˆCHC(CH3)2)
50
14
66
70
60
52
BuLi solution provided by Aldrich was used for the catalyst preparation.
ketones or propionates should provide a powerful tool for
the construction of two continuous chiral centers as well as
for the formation of carbon±carbon bonds. Catalytic asym-
metric syntheses of syn- and anti-aldols from latent enolates
have already been well investigated.1 In contrast, the
diastereo- and enantioselective synthesis of aldols, starting
from methylene ketones, by means of direct catalytic asym-
metric aldol reaction has not been reported.19 The bulkiness
of methylene ketones was anticipated to make it more dif®-
cult for the catalysts to abstract an a-hydrogen of the
ketones. We expected that the new catalysts could be applic-
able to diastereoselective direct aldol reactions.
The reaction between aldehyde 12 and 3-pentanone (13)
(Table 7) can produce the syn- or anti-aldol adduct (14),
which has been utilized as an intermediate in a total
synthesis of epothilone A by our group.20 First of all we
carried out the reaction in the presence of the hetero-
polymetallic catalyst (20 mol%); however, no reaction
proceeded and the starting materials were recovered. In
contrast, catalyst A was found to promote the aldol reaction
to afford the corresponding products, albeit in 15% yield
(entry 1). In order to improve the reactivity, we designed
a different catalyst (cat B), which possesses a tertiary
alkoxide instead of a primary alkoxide (ligand 15, Scheme
3). The catalyst (cat B) was prepared from La(O-i-Pr)3,
BuLi, and ligand 15 possessing tertiary alcohols. As a result,
catalyst B was found to improve the reactivity to give the
products in 38% yield (entry 2). The diastereoselectivity
was found to be opposite to that observed in the case of
catalyst A. Furthermore, asymmetric induction (28% ee)
has been achieved for the ®rst time by addition of LiI
(20 mol%) (entry 3). Although further addition of LiI
increased the ee (up to 51%), the chemical yield dropped
to 5% (entries 4, 5). The addition of other lithium salts gave
less satisfactory results (entries 6, 7). Unfortunately, catalyst
B was found to be less effective than catalyst A in the
reaction of aldehyde 7 with ketone 8a. The product (9a)
was obtained in only 24% ee and 33% yield (catalyst B,
20 mol%; 2308C; 90 min). When LiI was added, an
improvement on the solubility of the catalyst in toluene
was observed. This phenomenon implies the association of
LiI with the catalyst. However, its role is still not clear at the
moment.
Scheme 5. Postulated catalytic cycle.
intermediate II. The aldehyde replaces the lithium alkoxide
by coordination to the center metal to form intermediate III,
in which the aldehyde is activated by the Lewis acidity of
the center metal18 and the relative position of the aldehyde
and the enolate is arranged suitably. The enolate then
attacks the Si face of the aldehyde to afford the correspond-
ing aldolate (IV). As the last step protonation of the aldolate
by the alcohol moiety of the catalyst gives the (S)-aldol
adduct and regenerates the catalyst (I).§ The mechanism
of the enantioselection at the stage of intermediate III is
tentatively described in Fig. 2.
2.2. Attempts at diastereo- and enantioselective direct
catalytic aldol reaction of 3-pentanone
The aldol reaction between aldehydes and methylene
§
Another catalytic cycle could also be reasonable: starting from the cata-
lyst (I), ®rst coordination of the aldehyde could take place. After depro-
tonation of the ketone, again intermediate III could be formed.

2574
N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
Figure 2. Postulated intermediate. Several atoms and groups are omitted for clarity.
Table 7. Aldol reaction of 3-pentanone
Entry
Ligand
Additive
Yield (%)
syn/anti
ee of syn
ee of anti
Cf
1b
2
3
4
5
6
7
(Heteropolymetallic catalysta)
No reaction
15
38
21
11
5
24
±
±
±
±
6 (cat A)
15 (cat B)
15 (cat B)
15 (cat B)
15 (cat B)
15
LiI (10 mol %)
±
75/25
26/74
19/81
36/64
40/60
21/79
±
,10
21
8
10
9
24
28
36
51
26
±
LiI (20 mol %)
LiI (40 mol %)
LiI (60 mol %)
LiOH (20 mol %)
LiBF4 (20 mol %)
4
15
Trace
±
BuLi solution provided by Aldrich was used for the catalyst preparation.
The reaction was carried out in THF.
The reaction was carried out for 114 h.
a
b
3. Conclusions
Since this kind of reaction is totally unexploited, it is notable
that the anti-aldol product was obtained with up to 51% ee,
albeit in low chemical yield.
In summary, we have demonstrated a novel concept which
enables ¯exible design of Lewis acid±Brùnsted base bifunc-
tional catalysts. Lithium alkoxide was introduced as
Brùnsted base function to enhance the catalytic activity.
The lithium alkoxide moiety was covalently bonded to the
catalysts to prevent undesired background reactions, which
might form racemic products. In line with this concept,
several novel catalysts have been developed for direct asym-
metric aldol reactions. A catalyst possessing two primary
alkoxides was found to promote the aldol reactions of
dialkyl ketones as well as aromatic ketones, giving the
aldol products in moderate yield and ee, whereas a catalyst
lacking the alkoxide moieties gave no aldol products. On the
other hand, a different catalyst possessing tertiary alkoxide
moieties has been found to be applicable to the aldol reac-
tion of 3-pentanone, which should provide a powerful
method for the construction of continuous chiral centers.
Investigations into the role of lithium salts and further
studies on the development of diastereo- and enantio-
selective direct catalytic aldol reaction in accordance with
the methodology reported herein, for instance seeking a
more suitable Brùnsted base, are now in progress.23
4. Experimental
4.1. General methods and materials
Infrared (IR) spectra were recorded on a JASCO FT/IR-410
Fourier transform infrared spectrophotometer. NMR spectra
were recorded on a JEOL JNM-LA500 spectrometer,
1
operating at 500 MHz for H NMR and 125.65 MHz for

N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
2575
13C NMR. Chemical shifts in CDCl3 are reported on the d
scale relative to CHCl3 (7.26 ppm for 1H NMR and
77.00 ppm for 13C NMR) as an internal reference. The
following abbreviations are used to report multiplicities:
`s' (singlet), `d' (doulet), `dd' (doublet of doublets), `t'
(triplet), `dq' (doublet of quartets), `m' (multiplet), `br'
(broad). Optical rotations were measured on a JASCO
P-1010 polarimeter. EIMS were measured on a JEOL
JMS-BU20 GCmate. Column chromatography was carried
out with silica gel Merck 60 (230±400 mesh ASTM).
Preparative thin layer chromatography (preparative TLC)
was performed on Merck Art. 5715, Silica gel 60 F254 plates.
The enantiomeric excess (ee) was determined by HPLC
analysis. HPLC was performed on JASCO HPLC systems
consisting of the following: pump, PU-980; detector,
UVIDEC-100-IV, measured at 254 nm; column, DAICEL
CHIRALPAK AS, AD, DAICEL CHIRALCEL OD or OJ;
mobile phase, hexane-2-propanol; ¯ow rate, 0.3±1.0 mL/
min. Reactions were carried out in dry solvents under an
argon atmosphere. Tetrahydrofuran (THF) was distilled
from sodium benzophenone ketyl. Dichloromethane
(CH2Cl2) was distilled from calcium hydride. Toluene was
distilled from sodium. BuLi solution in hexane was
purchased from Kanto Chemical, Japan unless otherwise
stated. Lanthanum triisopropoxide (La(O-i-Pr)3) was
purchased from Kojundo Chemical Laboratory, 5-1-28,
Chiyoda, Sakado-shi, Saitama 350-0214, Japan (fax: 181-
492-84-1351). Other reagents were puri®ed by usual
methods.
7.19 (m, 2H), 7.14 (d, Jˆ8.0 Hz, 2H), 4.52 (d, Jˆ5.4 Hz,
2H), 4.39 (d, Jˆ5.4 Hz, 2H), 4.03±4.00 (m, 4H), 3.22 (dd,
Jˆ6.0, 13.5 Hz, 2H), 3.15 (dd, Jˆ6.5, 13.5 Hz, 2H), 2.97 (s,
6H), 2.12 (brs, 2H); 13C NMR (CDCl3) d 153.23, 133.18,
132.38, 130.86, 129.98, 127.51, 126.13, 125.91, 125.28,
125.12, 99.01, 63.23, 56.68, 34.45; FTIR (neat) n 3398,
2935, 1236, 1157, 1070 cm21; MS (m/z) 338 (base peak),
430, 462 (M1); [a]D ˆ2152.6 (c 0.97, CH2Cl2); HRMS
22
calcd for C28H30O6 462.2042, found 462.2040.
4.1.2. (R)-4-(3-(2-Allyloxyethyl)-2-methoxymethyloxy-1-
naphthyl)-3-methoxymethyloxy-2-naphthaleneethanol
(3). To a solution of 2 (19.4 g, 0.0419 mol) in acetonitrile
(210 mL), were added KF-alumina14 (25.6 g, containing
0.126 mol of KF) and allyl bromide (4.72 mL,
0.0545 mol). The resulting mixture was stirred vigorously
at room temperature for 20 h and ®ltered through a pad of
Celite. The ®ltrate was evaporated to dryness and the resi-
due was puri®ed by ¯ash silica gel column chromatography
(hexane/ethyl acetateˆ2/1) to afford allyl ether 3 (10.1 g, y.
48%) as a colorless oil; 1H NMR (CDCl3) d 7.86 (d,
Jˆ9.3 Hz, 2H), 7.82 (d, Jˆ7.9 Hz, 2H), 7.38±7.34 (m,
2H), 7.22±7.18 (m, 2H), 7.14±7.12 (m, 2H), 5.96±5.88
(m, 1H), 5.29±5.25 (m, 1H), 5.18±5.15 (m, 1H), 4.55 (d,
Jˆ5.4 Hz, 1H), 4.51 (d, Jˆ5.6 Hz, 1H), 4.41±4.38 (m, 2H),
4.05±4.00 (m, 4H), 3.88±3.78 (m, 2H), 3.28±3.13 (m, 4H),
2.96 (s, 3H), 2.90 (s, 3H), 2.15 (br, 1H); 13C NMR (CDCl3)
d 153.27, 153.12, 134.89, 133.27, 133.10, 132.46, 132.38,
130.87, 130.80, 129.93, 129.90, 127.52, 127.47, 126.04,
126.01, 125.99, 125.91, 125.5, 125.10, 125.07, 124.96,
116.81, 99.05, 98.92, 71.87, 70.30, 63.28, 56.68, 56.58,
34.45, 31.41; FTIR (neat) n 3446, 2933, 1158, 1070, 976,
922 cm21; MS (m/z) 338 (base peak), 426, 502 (M1);
4.1.1.
(R)-3,30-Bis(2-hydroxyethyl)-2,20-bis(methoxy-
methyloxy)-1,10-binaphthyl (2). To a solution of MOM-
protected BINOL 1 (20 g, 0.0534 mol) in 600 mL of
anhydrous ether, was added n-butyllithium (n-BuLi) in
hexane (110 mL, 0.171 mol, 1.56 N) at room temperature,
and the resulting mixture was stirred for 3.5 h. During the
stirring, an ether solution of ethylene oxide was prepared in
another ¯ask according to the procedure described below. A
dispersion of sodium hydride (21.4 g, 60±72% in mineral
oil) was suspended in 400 mL of anhydrous ether, and
2-chloroethanol (32.2 mL, 0.481 mol, distilled from
Na2SO4) was added slowly over 20 min with stirring at
08C. The resulting mixture was stirred at the same tempera-
ture for 2.5 h and the supernatant was used as ethylene oxide
solution. Tetrahydrofuran (200 mL) and the ether solution
of ethylene oxide prepared above were added to a cooled
suspension of the lithiated MOM-BINOL at 2968C
(acetone/liq.N2 bath). The mixture was stirred for 10 min
and boron tri¯uoride ethyl ether complex (BF3´OEt2)
(26.9 mL, 0.214 mol) was added in one portion. The stirring
was continued at 2968C for 25 min until the bath tempera-
ture was raised to 2788C. After stirring for 15 min, the
reaction mixture was poured into a mixture of saturated
aq. NaHCO3 (800 mL), water (800 mL) and ether
(800 mL). The aqueous layer was separated, saturated
with NaCl and extracted with ether (500 mL) and then
ethyl acetate (500 mL£5). The organic layers were
combined, washed with brine (500 mL£2) and dried over
Na2SO4. The solvent was removed under reduced pressure
and the residue was puri®ed by ¯ash silica gel column chro-
matography (hexane/ethyl acetateˆ1/1) to afford alcohol 2
(12.3 g, y. 50%) as a colorless oil; 1H NMR (CDCl3) d 7.85
(s, 2H), 7.82 (d, Jˆ7.8 Hz, 2H), 7.39±7.35 (m, 2H), 7.23±
25
[a]D ˆ2172.7 (c 2.27, CHCl3); HRMS calcd for
C31H34O6 502.2355, found 502.2357. Further elution with
hexane/acetone (1/1) gave the starting alcohol (2) (6.79 g,
35%).
4.1.3. (R)-4-(3-(2-Allyloxyethyl)-2-benzyloxy-1-naphthyl)-
3-benzyloxy-2-naphthaleneethanol (4). A mixture of allyl
ether 3 (26.4 g, 0.0525 mol) and p-toluenesulfonic acid
monohydrate (TsOH´H2O) (5.00 g, 0.0263 mol) in
CH2Cl2/MeOH (530/530 mL) was stirred at room tempera-
ture for 6.5 h. The resulting solution was poured into a
mixture of saturated aq. NaHCO3 (500 mL) and water
(500 mL). After stirring for 1 h, the aqueous layer was
separated and extracted with CH2Cl2 (300 mL£3). The
combined organic extracts were washed with saturated aq.
NaHCO3 (100 mL£2) and brine (200 mL) and dried over
Na2SO4. Removal of the solvent gave crude phenol, which
was used without further puri®cation for the next alkylation.
The crude phenol was dissolved in THF/DMF (250/250 mL)
and sodium hydride (2.52 g, 0.105 mol) was added to the
solution at 08C. The mixture was stirred at room temperature
for 30 min and then cooled to 08C. After the addition of
benzyl bromide (15.6 mL, 0.131 mol) the temperature was
allowed to raise to room temperature and the mixture was
stirred for 3.5 h. Water (600 mL) and ether (100 mL) were
added and the aqueous layer was extracted with ether
(200 mL£3). The combined organic layers were washed
with water (200 mL£2) and brine and dried over Na2SO4.
The solvent was removed under reduced pressure and the
resulting residue was puri®ed by ¯ash silica gel column

2576
N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
chromatography (hexane/ethyl acetateˆ3/1) to afford
benzyl ether 4 (22.8 g, y. 73%) as a colorless oil; 1H
NMR (CDCl3) d 7.89±7.83 (m, 4H), 7.41±7.36 (m, 2H),
7.27±7.20 (m, 4H), 7.15±7.07 (m, 6H), 6.72±6.68 (m, 4H),
5.93±5.81 (m, 1H), 5.26±5.21 (m, 1H), 5.15±5.12 (m, 1H),
4.63 (d, Jˆ10.8 Hz, 1H), 4.57 (d, Jˆ10.9 Hz, 1H), 4.30±
4.26 (m, 2H), 3.96±3.94 (m, 2H), 3.88±3.80 (m, 2H), 3.79±
3.73 (m, 1H), 3.70±3.64 (m, 1H), 3.23±3.09 (m, 3H), 3.03±
2.97 (m, 1H), 1.62 (br, 1H); 13C NMR (CDCl3) d 154.74,
154.57, 137.36, 137.13, 134.92, 133.50, 133.30, 132.63,
132.52, 130.87, 129.99, 129.88, 128.03, 127.97, 127.66,
127.65, 127.60, 127.52, 127.46, 126.05, 125.95, 125.82,
125.72, 125.57, 125.18, 124.98, 124.84, 116.77, 74.99,
74.94, 71.78, 70.50, 63.13, 34.87, 31.39; FTIR (neat) n
3423, 2931, 2866, 1235, 1101, 749 cm21; MS (m/z) 91
(hexane/acetoneˆ8/1!6/1) to afford alcohol 4 (1.30 g,
12%) and the symmetrical ether (7.58 g), which contained
inseparable impurity. A THF solution of zinc chloride
(ZnCl2) (37.2 mL, 0.5 M) and tetrakis(triphenylphosphine)-
palladium(0) (Pd(PPh3)4) (1.59 g, 1.38 mmol) were added to
a solution of the symmetrical ether (8.08 g) in THF
(35.6 mL). Under stirring, tributyltin hydride (11.7 mL,
43.5 mmol) was added slowly (over 1 h) to the mixture
before ethyl acetate (60 mL) and water (60 mL) were
added. The aqueous layer was separated and extracted
with ethyl acetate (20 mL£2). The combined organic layers
were washed with brine and dried over Na2SO4. After
removal of the solvent the residue was dissolved in
400 mL of acetonitrile and the solution was extracted with
hexane (100 mL£2). The acetonitrile layer was evaporated
to dryness and the resulting residue was puri®ed by ¯ash
silica gel column chromatography (hexane!hexane/
acetone 5/2, and then hexane/ethyl acetate 2/1!1/1) to
afford the desired alcohol (6.4 g) as a colorless foam,
which contained inseparable impurity. To a solution of the
alcohol (300 mg, 0.275 mmol) in THF/liquid ammonia
(3 mL/ca. 3 mL) was added lithium (22.9 mg, 3.30 mmol)
at 2788C. The resulting deep blue solution was stirred at
2788C for 13 min and the reaction was quenched by addi-
tion of ammonium chloride (2 g). After the evaporation of
ammonia, water and CH2Cl2 were added and the aqueous
layer was extracted with CH2Cl2. The combined organic
layers were washed with brine and dried over Na2SO4.
The solvent was removed under reduced pressure and the
residue was puri®ed by ¯ash silica gel column chromato-
graphy (CH2Cl2/acetone 15/1) to afford ligand 6 (234 mg, y.
70%) as a colorless powder. This sample contained 0.06%
(w/w) of CH2Cl2 and 5% (w/w) of acetone; 1H NMR
(CDCl3): d 7.76 (d, Jˆ7.9 Hz, 2H), 7.67 (d, Jˆ8.2 Hz,
4H), 7.63 (d, Jˆ7.9 Hz, 2H), 7.27±7.21 (m, 4H), 7.17±
7.12 (m, 4H), 7.03±7.00 (m, 4H), 6.70 (br, 2H), 6.22 (br,
2H), 3.91±3.74 (m, 8H), 3.16±3.04 (m, 6H), 2.97±2.91 (m,
2H); 13C NMR (CDCl3) d 151.82, 151.37, 132.95, 132.89,
130.64, 130.57, 129.15, 128.14, 127.96, 127.65, 126.42,
126.27, 124.43, 124.32, 123.72, 123.66, 113.82, 113.50,
71.65, 62.91, 34.76, 32.23; FTIR (KBr) n 3511, 3289,
2926, 1626, 1502, 1436, 1200, 1146, 1100, 750 cm21; MS
(Bn1, base peak), 445, 594 (M1); [a]D ˆ2159.6 (c 0.80,
23
CHCl3); HRMS calcd for C41H38O4 594.2770, found
594.2752.
4.1.4. (R)-4-(3-(2-Allyloxyethyl)-2-benzyloxy-1-naphthyl)-
3-benzyloxy-2-naphthaleneacetaldehyde (5). Benzyl
ether 4 (22.6 g, 0.0380 mol) was dissolved in CH2Cl2
(170 mL) and Dess±Martin periodinane (19.3 g, 0.0456
mol) was added at 08C. The resulting mixture was stirred
at room temperature for 2 h and then cooled to 08C. The
mixture was diluted with CH2Cl2 (200 mL) before saturated
aq. NaHCO3 (300 mL) and Na2S2O3´5H2O (20 g) were
added. The mixture was stirred at room temperature for
30 min and ®ltered through a pad of Celite. The ®ltrate
was evaporated to dryness and the residue was puri®ed by
¯ash silica gel column chromatography (hexane/ethyl
acetateˆ6/1) to afford aldehyde 5 (17.3 g, y. 77%) as a
1
colorless oil; H NMR (CDCl3) d 9.58±9.57 (m, 1H), 7.90
(s, 1H), 7.86±7.84 (m, 2H), 7.79 (s, 1H), 7.42±7.36 (m, 2H),
7.28±7.23 (m, 4H), 7.15±7.04 (m, 6H), 6.71±6.69 (m, 4H),
5.93±5.84 (m, 1H), 5.26±5.21 (m, 1H), 5.15±5.12 (m, 1H),
4.61 (d, Jˆ10.8 Hz, 1H), 4.43 (d, Jˆ10.6 Hz, 1H), 4.27 (d,
Jˆ10.6 Hz, 1H), 4.26 (d, Jˆ10.8 Hz, 1H), 3.96±3.94 (m,
2H), 3.84±3.65 (m, 4H), 3.23±3.16 (m, 1H), 3.15±3.08 (m,
1H); 13C NMR (CDCl3) d 199.68, 154.81, 154.19, 137.20,
136.75, 134.91, 134.08, 133.22, 132.70, 130.88, 130.83,
130.74, 130.10, 128.09, 128.00, 127.78, 127.76, 127.72,
127.58, 127.53, 126.74, 126.58, 126.10, 125.89, 125.66,
125.57, 125.21, 124.94, 124.79, 116.80, 75.08, 74.89,
71.80, 70.46, 45.94, 31.38; FTIR (neat) n 2859, 2726,
1723, 1235, 1102, 751 cm21; MS (m/z) 91 (Bn1, base
(m/z) 712 (M12H2O), 730 (M1); [a]D ˆ123.9 (c 1.13,
22
CHCl3); HRMS calcd for C48H42O7 730.2931, found
730.2921.
peak), 443, 592 (M1); [a]D ˆ2161.4 (c 1.20, CHCl3);
4.1.6. (R,R)-3,300-Oxydiethylene-di-(2,20-dihydroxy-30-
(2-hydroxy-2,2-dimethylethyl)-1,10-binaphthyl) (15). To
a solution of the alcohol which was obtained after the depro-
tection of allyl ethers in the synthesis of ligand 6 (608 mg,
0.557 mmol), in CH2Cl2 (2.9 mL), was added Dess±Martin
periodinane (543 mg, 1.28 mmol) at 08C. The resulting
mixture was stirred at room temperature for 5 h. The reac-
tion was quenched by the addition of Na2S2O3´H2O
(500 mg), saturated aq. NaHCO3 (5 mL) and CH2Cl2.
After stirring for 1 h, the mixture was ®ltered through a
pad of Celite. The aqueous layer was separated and
extracted with CH2Cl2. The combined organic layers were
washed with saturated aq. NaHCO3 and brine and dried over
Na2SO4. The solvent was evaporated to dryness and the
residue was puri®ed by ¯ash silica gel column chromato-
graphy (hexane/acetone 4/1) to afford the aldehyde
(496 mg, y. 82%). The aldehyde was dissolved in THF
22
HRMS calcd for C41H36O4 592.2613, found 592.2639.
4.1.5. (R,R)-3,300-Oxydiethylene-di-(2,20-dihydroxy-30-
(2-hydroxyethyl)-1,10-binaphthyl) (6). To a cooled solu-
tion of triethylsilane (Et3SiH) (6.2 mL, 39.0 mmol) in
CH2Cl2 (40 mL) were added trimethylsilyl tri¯uoro-
methanesulfonate (TMSOTf) (0.336 mL, 1.86 mmol) and
a solution of aldehyde 5 (11.0 g, 18.6 mmol) in CH2Cl2
(30 mL110 mL rinse) at 08C. After stirring for 1 h, the
reaction was quenched by addition of saturated aq.
NaHCO3 (40 mL) and CH2Cl2 (40 mL) at 08C. The aqueous
layer was separated and extracted with CH2Cl2 (20 mL£3).
The combined organic layers were washed with saturated
aq. NaHCO3 (30 mL) and brine and dried over Na2SO4. The
solvent was removed under reduced pressure and the residue
was puri®ed by ¯ash silica gel column chromatography

N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
2577
(6 mL) and the solution was added to a cooled solution of
methylmagnesium bromide (1.00 mmol) in THF (3 mL) at
08C. The mixture was stirred at 08C for 15 min and
quenched by adding aq. HCl (1N, 5 mL). The mixture was
extracted with CH2Cl2, and the extract was washed with
brine and dried over Na2SO4. The solvent was removed
under reduced pressure and the resulting residue was puri-
®ed by ¯ash silica gel column chromatography (hexane/
ethyl acetate 2/1!1/1) to afford the secondary alcohol as
a mixture of diastereomers (325 mg, y. 73%). This alcohol
(325 mg) was then treated with Dess±Martin periodinane
(283 mg) in CH2Cl2 (5.8 mL) at 08C, and the stirring was
continued for 4 h until saturated aq. NaHCO3 was added.
The mixture was ®ltered and the aqueous layer was
extracted with CH2Cl2. The extract was washed with satu-
rated aq. NaHCO3 and brine and dried over Na2SO4. The
solvent was removed under reduced pressure and the result-
ing residue was puri®ed by ¯ash silica gel column chroma-
tography (hexane/ethyl acetate 3/1) to afford the desired
ketone (262 mg, y. 81%) (This ketone was very unstable
and should be subjected to the next reaction immediately).
Cerium chloride21 (anhydrous, 326 mg, 0.325 mmol) was
dried under reduced pressure for 40 min at 908C and then
for 3 h at 1408C. After suspending cerium chloride in THF
(800 mL), it was sonicated for 1 h. The resulting slurry was
vigorously stirred for 11 h. A solution of methyllithium
(1.16 mL, 1.32 mmol, 1.14 M in ether) was added at
2408C and the resulting mixture was stirred at the same
temperature for 2.5 h. The mixture was cooled to 2788C
and a solution of the ketone (262 mg) in THF (1.5 mL)
was added 2788C. The stirring was continued at 2408C
for 10 min until the reaction was quenched by addition of
saturated aq. NH4Cl (3 mL). The aqueous layer was sepa-
rated and extracted twice with CH2Cl2. The combined
organic layers were washed with water (twice) and brine
and dried over Na2SO4. The solvent was removed under
reduced pressure and the resulting residue was puri®ed by
¯ash silica gel column chromatography (hexane/ethyl
acetateˆ3/1) to afford the tertiary alcohol (233 mg). The
tertiary alcohol (1.01 g) was dissolved in THF (10 mL)/
liq. NH3 (ca. 10 mL) in a cooled ¯ask (2788C). Lithium
(73.3 mg) was added and the resulting blue solution was
stirred at 2788C for 10 min. Ammonium chloride (6 g)
was added to quench the reaction and ammonia was allowed
to evaporate. Dichloromethane and water were added to the
mixture and the aqueous layer was extracted with CH2Cl2.
The combined organic layers were washed with brine and
dried over Na2SO4. After evaporation of the solvent, puri-
®cation was performed by ¯ash silica gel column chromato-
graphy (hexane/ethyl acetate and then hexane/acetone) to
afford ligand 15 (400 mg) as a pale yellow powder. This
sample contained 3% (w/w) of diethyl ether; 1H NMR
(CDCl3): d 7.80 (d, Jˆ7.9 Hz, 2H), 7.71±7.62 (m, 6H),
7.32±7.28 (m, 2H), 7.24±7.13 (m, 6H), 7.09±7.02 (m,
4H), 3.93±3.84 (m, 4H), 3.20±3.03 (m, 8H), 1.30 (s, 6H),
1.24 (s, 6H); 13C NMR (CDCl3): d 151.94, 151.25, 133.28,
132.93, 132.07, 130.20, 129.17, 129.03, 128.15, 127.69,
127.36, 126.33, 126.14, 124.67, 124.49, 123.61, 123.48,
114.95, 114.73, 72.85, 71.93, 45.44, 32.34, 29.79, 29.57;
FTIR (KBr) n 3520, 3229, 2970, 1626, 1501, 1435, 1203,
1099, 749 cm21; MS (m/z) 750 (M1 - 2H2O), 768 (M1-
4.2. General procedure for the aldol reactions promoted
by catalyst A
1
Ligand 6 (27.9 mg, calculated on the basis of H NMR;
0.0381 mmol), which contained trace amount of diethyl
ether and acetone, was dissolved in toluene/CH2Cl2. The
solvents were evaporated together with trace amount of
diethyl ether and acetone at room temperature under
reduced pressure. The residue was dried at room tempera-
ture under reduced pressure for 1 h and dissolved in THF
(200 mL). This solution was cooled to 08C, and a solution of
lithium iodide (0.0381 mmol, 0.5 M in THF) was added. A
solution of lanthanum triisopropoxide (La(O-i-Pr)3)
(0.0381 mmol, 0.2 M) in THF was added at 08C and the
resulting mixture was stirred for 40 min at the same
temperature. Addition of n-BuLi (0.114 mmol, 1.62N in
hexanes, Aldrich) to this solution was followed by stirring
for 30 min at 08C, and the solvent was evaporated to dryness
at the same temperature under reduced pressure. Toluene
(200 mL) was added to the residue and the resulting mixture
was stirred at room temperature for 10 min. The solvent was
evaporated and the residue was dried under reduced pres-
sure at room temperature for 10 min. Toluene (300 mL) was
again added to the residue and the resulting mixture was
cooled to 230 or 2208C (see Table 6). The ketone (8a±
8d) (2±5 mol equiv. with respect to the aldehyde
(0.191 mmol), see Table 6) was added to the stirred mixture
of the catalyst. After stirring for 20 min, 2,2-dimethyl-3-
phenylpropanal (7) (32 mL, 0.191 mmol) was added and
the resulting mixture was stirred at 230 or 2208C (see
Table 6). The reaction was quenched by addition of aq.
HCl (1N, 2 mL) and CH2Cl2 (2 mL). The aqueous layer
was separated and extracted twice with CH2Cl2. The organic
layers were combined, washed with brine and dried over
Na2SO4. The solvent was evaporated under reduced pres-
sure and the residue was puri®ed by ¯ash silica gel column
chromatography to afford the corresponding product (9a±
9d). The ligand (6) was recovered by elution with CH2Cl2/
acetone. The enantiomeric excess was determined by HPLC
analysis. Aldol products 9a and 9b have been reported (Ref.
9). The absolute con®gurations of 9c and 9d were deter-
1
mined to be S on the basis of H NMR analysis of the
corresponding MTPA esters.22
4.2.1. (S)-3-Hydroxy-4,4-dimethyl-1,5-diphenyl-1-penta-
none (9a, Table 6, entry 1). Yield 70%, 67% ee; mixture of
colorless solid and colorless oil; ¯ash column (SiO2):
hexane/ether 15/1; HPLC: DAICEL CHIRALPAK AD,
2-propanol/hexane 1/9, ¯ow 1.0, tR 8.4 min (S) and
10.2 min (R).
4.2.2. (S)-6,6-Dimethyl-5-hydroxy-7-phenyl-3-heptanone
(9b, Table 6, entry 2). Yield 71%, 40% ee; colorless oil;
¯ash column (SiO2): hexane/ether 10/1; HPLC: DAICEL
CHIRALPAK AD, 2-propanol/hexane 5/95, ¯ow 0.2, tR
37.2 min (R) and 42.6 min (S).
4.2.3. (S)-5-Hydroxy-2,6,6-trimethyl-7-phenyl-3-hepta-
none (9c, Table 6, entry 3). Yield 62%, 45% ee; colorless
oil; ¯ash column (SiO2): hexane/ether 15/1; 1H NMR
(CDCl3) d 7.26±7.16 (m, 5H), 3.73 (dd, Jˆ1.5, 10.0 Hz,
1H), 3.22 (brs, 1H), 2.79 (d, Jˆ12.3 Hz, 1H), 2.66 (dd,
Jˆ1.5, 16.8 Hz, 1H), 2.63±2.55 (m, 1H), 2.51 (dd,
H2O), 786 (M1); [a]D ˆ135.6 (c 0.60, CHCl3); HRMS
23
calcd for C48H42O7 786.3557, found 786.3572.

2578
N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
Jˆ10.0, 16.8 Hz, 1H), 2.48 (d, Jˆ12.3 Hz, 1H), 1.09 (d,
Jˆ6.7 Hz, 3H), 1.03 (d, Jˆ6.6 Hz, 3H), 0.89 (s, 3H), 0.79
(s, 3H); 13C NMR (CDCl3) d 216.80, 138.68, 130.81,
127.74, 125.90, 72.94, 44.50, 41.68, 41.33, 37.90, 23.33,
22.25, 18.11, 18.06; FTIR (neat) n 3502, 2967, 2930,
1703, 1081 cm21; MS (m/z) 91 (Bn1, base peak), 115, 230
NMR analysis of the corresponding MTPA ester.22 (^)-
Anti-14 was reported in Ref. 20.
4.3.1. (4R,5S)-7-Benzyloxy-5-hydroxy-4,6,6-trimethyl-3-
heptanone (anti-14). Colorless oil; puri®cation: ¯ash
column (SiO2), hexane/ethyl acetateˆ15/1 then preparative
TLC, hexane/etherˆ4/1; 1H NMR (CDCl3) d 7.35±7.26 (m,
5H), 4.44 (d, Jˆ7.9 Hz, 1H), 4.43 (s, 2H), 3.44 (dd, Jˆ2.8,
7.9 Hz, 1H), 3.2623.21 (m, 2H), 2.87 (dq, Jˆ2.8, 7.0 Hz,
1H), 2.50 (dq, Jˆ7.0, 18.0 Hz, 1H), 2.44 (dq, Jˆ7.0,
18.0 Hz, 1H), 1.26 (d, Jˆ7.0 Hz, 3H), 0.95 (t, Jˆ7.0 Hz,
3H), 0.91 (s, 3H), 0.90 (s, 3H); 13C NMR (CDCl3) d
218.53, 138.31, 128.32, 127.55, 127.51, 81.66, 77.22,
73.17, 44.31, 39.87, 35.94, 22.90, 21.50, 17.94, 7.29;
FTIR (neat) n 3462, 2972, 2935, 2875, 1964, 1455,
1102 cm21; MS (m/z) 91 (Bn1), 149, 167 (base peak), 279
(M12H2O), 248 (M1); [a]D ˆ210.9 (c 0.4, CHCl3) (40%
22
ee); HRMS calcd for C17H26O3 248.1776, found 248.1777;
HPLC: DAICEL CHIRALPAK AD, 2-propanol/hexane 2/
98, ¯ow 0.3, tR 32.8 min (R) and 35.5 min (S).
4.2.4. (S)-6-Hydroxy-2,7,7-trimethyl-8-phenyl-2-octen-
4-one (9d, Table 6, entry 4). Yield 66%, 60% ee; colorless
solid; ¯ash column (SiO2): hexane/ether 15/1; H NMR
1
(CDCl3) d 7.26±7.23 (m, 2H), 7.19±7.14 (m, 3H), 6.06±
6.05 (m, 1H), 3.76 (d, Jˆ10.1 Hz, 1H), 3.42 (brs, 1H), 2.80
(d, Jˆ12.3 Hz, 1H), 2.62 (dd, Jˆ1.7, 16.7 Hz, 1H), 2.49 (dd,
Jˆ10.1, 16.7 Hz, 1H), 2.47 (d, Jˆ12.3 Hz, 1H), 2.14 (d,
Jˆ0.8 Hz, 3H), 1.88 (d, Jˆ0.8 Hz, 3H), 0.89 (s, 3H), 0.79
(s, 3H); 13C NMR (CDCl3) d 202.10, 156.80, 138.79,
130.85, 127.69, 125.83, 124.02, 73.11, 44.80, 44.50,
37.86, 27.75, 23.30, 22.28, 20.93; FTIR (KBr) n 3458,
2969, 2929, 1687, 1620 cm21; MS (m/z) 83 (base peak),
(MH1); [a]D ˆ14.5 (c 0.28, CHCl3); HRMS calcd for
22
C17H26O3 278.1882, found 278.1883; HPLC: CHIRALCEL
OJ, 2-propanol/hexane 2/98, ¯ow 0.5, tR 15.3 min (5S) and
17.1 min (5R).
4.3.2. (4Rp,5Rp)-7-Benzyloxy-5-hydroxy-4,6,6-trimethyl-
3-heptanone (syn-14). Colorless oil; puri®cation: ¯ash
column (SiO2), hexane/ethyl acetateˆ15/1 then preparative
TLC, hexane/ethyl acetateˆ4/1;1H NMR (CDCl3) d 7.36±
7.27 (m, 5H), 4.49 (s, 2H), 3.89±3.87 (m, 1H), 3.36 (d,
Jˆ8.8 Hz, 1H), 3.28 (d, Jˆ8.8 Hz, 1H), 3.26 (d,
Jˆ4.1 Hz, 1H), 2.75 (dq, Jˆ4.1, 7.0 Hz, 1H), 2.53 (dq,
Jˆ7.1, 17.8 Hz, 1H), 2.47 (dq, Jˆ7.1, 17.8 Hz, 1H), 1.15
(d, Jˆ7.0 Hz, 3H), 1.03 (t, Jˆ7.1 Hz, 3H), 0.93 (s, 3H), 0.91
(s, 3H); 13C NMR (CDCl3) d 215.11, 137.88, 128.43,
127.74, 127.57, 79.69, 76.17, 73.59, 47.64, 39.19, 34.20,
22.97, 20.79, 12.46, 7.81; FTIR (neat) n 3495, 2971,
2936, 2875, 1709, 1455 cm21; MS (m/z) 91 (Bn1, base
peak), 149, 167, 279 (MH1); HRMS calcd for C17H26O3
278.1882, found 278.1893; HPLC: CHIRALPAK AS,
2-propanol/hexane 1/9, ¯ow 0.5, tR 9.8 and 11.6 min.
91 (Bn1), 127, 242 (M1-H2O), 260 (M1); [a]D ˆ27.8 (c
24
0.65, CHCl3) (60% ee); HRMS calcd for C17H24O2
260.1776, found 260.1764; HPLC: CHIRALPAK AD,
2-propanol/hexane 5/95, ¯ow 1.0, tR 7.2 min (S) and
11.2 min (R).
4.3. Procedure for the aldol reaction promoted by
catalyst B (Table 7)
Ligand 15 (30 mg; 0.0381 mmol, containing 1 mg of diethyl
ether on the basis of 1H NMR) was dissolved in toluene. The
solution was evaporated to remove diethyl ether at room
temperature under reduced pressure. The residue was
dried at room temperature under reduced pressure for 1 h
and dissolved in toluene (300 mL). The solution was cooled
to 08C and a solution of lithium iodide (0.0381 mmol, 0.5 M
in THF) was added. A solution of La(O-i-Pr)3 (191 mL,
0.0381 mmol, 0.2 M in toluene) was added and the resulting
mixture was stirred at 08C for 30 min. The solvent was
removed at 08C under reduced pressure. After drying at
08C for 40 min, toluene (0.45±1 mL) was added to the
resulting residue. The mixture was stirred for 15 min and
n-BuLi (71 mL, 0.114 mmol, 1.62N in hexanes) was added
at 08C. After stirring at 08C for 30 min, the mixture was
cooled to 2208C. 3-Pentanone (13) (101 mL, 0.953 mmol)
was added and the mixture was stirred at 2208C for 20 min.
After the addition of 3-benzyloxy-2,2-dimethylpropanal
(12) (37 mL, 0.191 mmol), the mixture was stirred at
2208C for 141 h. The reaction was quenched by addition
of 1N aq. HCl (2 mL) and ether (2 mL). The aqueous layer
was separated and extracted twice with ether. The organic
layers were combined and washed with brine. After drying
over Na2SO4, the solvent was evaporated under reduced
pressure and the residue was puri®ed by ¯ash silica gel
column chromatography to afford syn-14 (2.1 mg, y. 4%)
and anti-14 (9 mg, y. 17%). Ligand 15 was eluted with
hexane/acetone. The ee of each product was determined
by HPLC analysis. The absolute con®guration of anti-14
Acknowledgements
This study was ®nancially supported by CREST, The Japan
Science and Technology Corporation (JST), and by a Grant-
in-Aid for Scienti®c Research from the Ministry of Educa-
tion, Science, and Culture, Japan. N. Yoshikawa thanks the
Japan Society for the Promotion of Science (JSPS) for a
research fellowship.
References
1. For recent examples, see: (a) Ishitani, H.; Yamashita, Y.;
Shimizu, H.; Kobayashi, S. J. Am. Chem. Soc. 2000, 122,
5403±5404. (b) Denmark, S. E.; Stavenger, R. A.; Wong,
K.-T.; Su, X. J. Am. Chem. Soc. 1999, 121, 4982±4991.
(c) Yanagisawa, A.; Matsumoto, Y.; Asakawa, K.;
Yamamoto, H. J. Am. Chem. Soc. 1999, 121, 892±893.
(d) Evans, D. A.; Burgey, C. S.; Kozlowski, M. C.; Tregay,
S. W. J. Am. Chem. Soc. 1999, 121, 686±699. (e) Evans, D. A.;
Kozlowski, M. C.; Murry, J. A.; Burgey, C. S.; Campos, K. R.;
Connell, B. T.; Staples, R. J. J. Am. Chem. Soc. 1999, 121,
669±685. (f) KruÈger, J.; Carreira, E. M. J. Am. Chem. Soc.
1998, 120, 837±838. (g) Chen, C.-T.; Chao, S.-D.; Yen, K.-C.;
1
was determined by Mosher's method on the basis of H

N. Yoshikawa, M. Shibasaki / Tetrahedron 57 (2001) 2569±2579
2579
Chen, C.-H.; Chou, I.-C.; Hon, S.-W. J. Am. Chem. Soc. 1997,
119, 11341±11342. (h). Sodeoka, M.; Tokunoh, R.; Miyazaki,
F.; Hagiwara, E.; Shibasaki, M. Synlett 1997, 463±466.
(i). Uotsu, K.; Sasai, H.; Shibasaki, M. Tetrahedron:
Asymmetry 1995, 6, 71±74. (j). Keck, G. E.; Krishnamurthy,
D. J. Am. Chem. Soc. 1995, 117, 2363±2364. (k). Mikami, K.;
Matsukawa, S. J. Am. Chem. Soc. 1994, 116, 4077±4078. (l).
Kobayashi, S.; Uchiro, H.; Shiina, I.; Mukaiyama, T.
Tetrahedron 1993, 49, 1761±1772. (m). Corey, E. J.;
Cywin, C. L.; Roper, T. D. Tetrahedron Lett. 1992, 33,
6907±6910. (n). Kiyooka, S.; Kaneko, Y.; Kume, K. Tetrahe-
dron Lett. 1992, 33, 4927±4930. (o). Parmee, E. R.; Hong, Y.;
Tempkin, O.; Masamune, S. Tetrahedron Lett. 1992, 33,
1729±1732.
see: (c) Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39,
1615.
11. For the development of Lewis acid-Lewis base bifunctional
catalysts, a phosphine oxide moiety has been covalently
attached to binaphthyl. See: (a) Takamura, M.; Funabashi,
K.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc. 2000, 122,
6327±6328. (b) Hamashima, Y.; Sawada, D.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 2641±2642
(and references cited therein).
12. Linked BINOLs have been prepared by our group and
Kobayashi's group. For oxygen-linked BINOL, see:
(a) Kim, Y. S.; Matsunaga, S.; Das, J.; Sekine, A.; Ohshima,
T.; Shibasaki, M. J. Am. Chem. Soc. 2000, 122, 6506±6507.
(b) Matsunaga, S.; Das, J.; Roels, J.; Vogl, E. M.; Yamamoto,
N.; Iida, T.; Yamaguchi, K.; Shibasaki, M. J. Am. Chem. Soc.
2000, 122, 2252±2260. For carbon-linked BINOLs, see:
(c) Vogl, E. M.; Matsunaga, S.; Kanai, M.; Iida, T.; Shibasaki,
M. Tetrahedron Lett. 1998, 39, 7917±7920. (d) Ishitani, H.;
Kitazawa, T.; Kobayashi, S. Tetrahedron Lett. 1999, 40, 2161.
13. (a) Eis, M. J.; Wrobel, J. E.; Ganem, B. J. Am. Chem. Soc.
1984, 106, 3693±3694. (b) Cox, P. J.; Wang, W.; Snieckus, V.
Tetrahedron Lett. 1992, 33, 2253±2256.
(p). Furuta, K.; Maruyama, T.; Yamamoto, H. J. Am. Chem.
Soc. 1991, 113, 1041±1042 (and references cited therein). For
some reviews, see: (q) Machajewski, T. D.; Wong, C.-H.
Angew. Chem., Int. Ed. 2000, 39, 1352±1374. (r). Nelson,
È
S. G. Tetrahedron: Asymmetry 1998, 9, 357±389. (s). Groger,
H.; Vogl, E. M.; Shibasaki, M. Chem. Eur. J. 1998, 4, 1137±
1141.
2. For a partially successful attempt, see: Nakagawa, M.; Nakao,
H.; Watanabe, K.-I. Chem. Lett. 1985, 391±394.
14. Ando, T.; Yamawaki, J.; Kawate, T.; Sumi, S.; Hanafusa, T.
Bull. Chem. Soc. Jpn 1982, 55, 2504±2507.
3. For catalytic asymmetric aldol reactions using catalytic anti-
bodies, see: (a) Tanaka, F.; Lerner, R. A.; Barbas III, C. F.
J. Am. Chem. Soc. 2000, 122, 4835±4836. (b) Hoffmann, T.;
Zhong, G.; List, B.; Shabat, D.; Anderson, J.; Gramatikova, S.;
Lerner, R. A.; Barbas III, C. F. J. Am. Chem. Soc. 1998, 120,
2768±2779 (and references cited therein).
15. Sassaman, M. B.; Kotian, K. D.; Prakash, G. K. S.; Olah, G. A.
J. Org. Chem. 1987, 52, 4314±4319.
16. Yin, H.; Franck, R. W.; Chen, S.-L.; Quigley, G. J.; Todaro, L.
J. Org. Chem. 1992, 57, 644±651.
17. Interesting effects of lithium halides on enantioselective reac-
tions were reported. See: Sugasawa, K.; Shindo, M.; Noguchi,
H.; Koga, K. Tetrahedron Lett. 1996, 37, 7377±7380. See also
Ref. 12b.
4. Catalytic asymmetric aldol reactions of a-isocyano carbonyl
compounds were reported. See: Sawamura, M.; Nakayama,
Y.; Kato, T.; Ito, Y. J. Org. Chem. 1995, 60, 1727±1732.
5. Recently, catalytic asymmetric cyclocondensations between
acyl halide and aldehyde have been reported. See: (a) Nelson,
S. G.; Peelen, T. J.; Wan, Z. J. Am. Chem. Soc. 1999, 121,
9742±9743. Another group has reported a rhodium-catalyzed
enantioselective reductive aldol reaction. See: (b) Taylor, S. J.;
Duffey, M. O.; Morken, J. P. J. Am. Chem. Soc. 2000, 122,
4528±4529.
18. In previous reports the coordination of the aldehyde to La was
strongly supported by spectroscopic experiments. See Ref. 7±
9.
19. The diastereo- and enantioselective aldol reaction of cyclo-
pentanone catalyzed by the heteropolymetallic catalyst was
reported. See Ref. 9. For catalytic antibody-catalyzed aldol
reactions, hydroxyacetone, ¯uoroacetone, and several kinds
of cyclic ketones have been used as substrates. See Ref. 3b.
For proline-catalyzed aldol reactions, hydroxyacetone has
been used as substrate. See Ref. 10a.
6. For a review: Shibasaki, M.; Sasai, H.; Arai, T. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1236±1256.
7. (a) Yamada, Y. M. A.; Yoshikawa, N.; Sasai, H.; Shibasaki,
M. Angew. Chem., Int. Ed. Engl. 1997, 36, 1871±1873. (b) C
& EN News 1997, September 8, 30
20. Sawada, D.; Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39,
209±213.
21. Bartoli, G.; Sambri, L.; Marcantoni, E.; Petrini, M. Tetrahedron
Lett. 1994, 35, 8453±8456.
8. Yamada, Y. M. A.; Shibasaki, M. Tetrahedron Lett. 1998, 39,
5561±5564.
22. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512±
519.
9. Yoshikawa, N.; Yamada, Y. M. A.; Das, J.; Sasai, H.;
Shibasaki, M. J. Am. Chem. Soc. 1999, 121, 4168±4178.
10. (a) Notz, W.; List, B. J. Am. Chem. Soc. 2000, 122, 7386±
7387. (b) List, B.; Lerner, R. A.; Barbas III, C. F. J. Am. Chem.
Soc. 2000, 122, 2395±2396. For an intramolecular reaction,
23. Note added in proof. While this work was reviewed, the
following communication describing a direct catalytic asym-
metric aldol reaction has been published. See: Trost, B. M.;
Ito, H. J. Am. Chem. Soc. 2000, 122, 12003±12004.