Enantioselective double aldol reaction catalyzed by chiral phosphine oxide

Article abstract of DOI:10.1002/chem.201100917

Goes on twice! The first enantioselective double aldol reaction is described (see scheme; R¹=aryl, alkyl, alkenyl; R²=aryl, alkenyl). A combination of readily available chiral phosphine oxide and silicon tetrachloride as reagents enables the unique transformation, which gives the double aldol adducts in high yields and selectivities.

Full text of DOI:10.1002/chem.201100917

DOI: 10.1002/chem.201100917
Enantioselective Double Aldol Reaction Catalyzed by Chiral Phosphine
Oxide
Yasushi Shimoda,[a] Shunsuke Kotani,*[b] Masaharu Sugiura,[a] and Makoto Nakajima*[a]
Sequential reactions have gained importance in recent
years because they provide attractive approaches to the syn-
thesis of molecules with complex architectures in a single
operational step.[1] One of the current challenges in organic
chemistry is to develop tandem reactions (domino or cas-
cade reactions) that provide complex molecules from readily
available starting compounds. Among the tandem reactions
identified, a few sequential aldol reaction strategies have
been demonstrated using boron,[2] silicon,[3] or metal re-
agents.[4] Over the past decade, direct aldol reactions have
attracted attention with the goal of improving the atom
economy.[5,6] The development of sequential enantioselective
direct aldol reactions would improve the preparation of
complex structures in the field of organic synthesis. Howev-
er, no reports have described the use of sequential enantio-
selective aldol reactions.
plied to this system, providing a unique double aldol adduct
in good yield with good diastereo- and enantioselectivity.
This approach constitutes the first enantioselective double
aldol reaction using silicon tetrachloride and a chiral phos-
phine oxide as a Lewis base organocatalyst.[9]
Our synthetic study began with the asymmetric double
aldol reaction of acetophenone (1a) and benzaldehyde (2a)
using chiral phosphine oxide, (S)-BINAPO (Scheme 2), at
À408C in dichloromethane (Table 1, entry 1). The reaction
was complete within 24 h and gave the corresponding prod-
uct 3a in 90% yield as a diastereomeric mixture (chiro/
We recently found that the combination of silicon tetra-
chloride and a phosphine oxide enables an aldol transforma-
tion through an enolization–aldol reaction process in situ
and developed an enantioselective direct aldol reaction of
cyclic ketones, catalyzed by
a chiral phosphine oxide
(Scheme 1).[7,8] The acyclic ketone acetophenone was ap-
Scheme 2. Chiral Lewis base catalysts used in this study.
Table 1. Optimization for the double aldol reaction of 1a and 2a.[a]
Scheme 1. Direct aldol and double aldol reactions using silicon tetrachlor-
ide and phosphine oxide.
Entry
Solvent[b]
Amine
Yield
[%][c]
d.r[d]
ee [%]
ACHTUNGTRENNUNG
(chiro)[e]
[a] Y. Shimoda, Prof. Dr. M. Sugiura, Prof. Dr. M. Nakajima
Graduate School of Pharmaceutical Sciences
Kumamoto University, 5–1 Oe-honmachi
Kumamoto 862-0973 (Japan)
1[f]
2
3
CH2Cl2
EtCN
EtCN/CH2Cl2
EtCN/CH2Cl2
EtCN/CH2Cl2
EtCN/CH2Cl2
iPr2NEt
iPr2NEt
iPr2NEt
pempidine
cHex2NMe
cHex2NMe
90
12
82
72
95
86
81:19
73:27
77:23
77:23
79:21
78:22
60
70
62
64
61
70
Fax : (+81)96-362-7692
E-mail: nakajima@gpo.kumamoto-u.ac.jp
4
5
[b] Prof. Dr. S. Kotani
6[g]
Priority Organization for Innovation and Excellence
Kumamoto University, 5–1 Oe-honmachi
Kumamoto 862-0973 (Japan)
Fax : (+81)96-371-4394
E-mail: skotani@gpo.kumamoto-u.ac.jp
[a] Unless otherwise noted, reactions were carried out by addition of sili-
con tetrachloride (1.0 mmol) to a solution of ketone 1a (0.25 mmol), al-
dehyde 2a (0.55 mmol), an amine (1.25 mmol), and (S)-BINAPO (10
mol%) in solvent (2.5 mL) at À408C. [b] EtCN/CH2Cl2 =1:1. [c] Yield of
the diastereomers. [d] Determined by 1H NMR spectroscopy (chiro/
meso). [e] Determined by HPLC analysis. [f] For 8 h. [g] At À608C.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201100917.
7992
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7992 – 7995

COMMUNICATION
meso=81:19).[10] The chiro product (chiro-3a) was obtained
with 60% ee.[11] A variety of solvents was tested to improve
the catalytic activity and selectivity. Relatively high enantio-
selectivity was observed in propionitrile, although the reac-
tivity was even higher in dichloromethane.[12] The use of
EtCN/CH2Cl2 (1:1) as a mixed solvent slightly increased the
enantioselectivity without loss of the reactivity (Table 1,
entry 3).
Table 3. Enantioselective double aldol reaction of various ketones 1 with
2a.[a]
Entry Ketone 1
R
Product 3 Yield d.r.[c]
ee [%]
ACHTUNGTRENNUNG
(chiro)[d]
[%][b]
We next studied the effect of amines on the reaction of
1a with 2a, catalyzed by (S)-BINAPO in EtCN/CH2Cl2
(1:1). Sterically congested aliphatic amines gave good results
(Table 1, entries 4 and 5).[13] In particular, dicyclohexylme-
thylamine allowed the reaction to proceed smoothly, afford-
ing the product in good yield and selectivity (Table 1,
entry 5). Moreover, lowering the reaction temperature to
À608C improved the enantioselectivity, yielding the product
3a with 70% ee (Table 1, entry 6).
Next, we tested various chiral Lewis bases in the double
aldol reaction of 1a and 2a at À608C (Scheme 2). (S)-tol-
BINAPO gave a chemical yield and selectivity similar to
those obtained with (S)-BINAPO (Table 2, entries 1 and 2).
1
1a
1b
1c
1d
1e
1 f
1g
1h
1i
Ph
4-BrC6H4
3a
3b
86
78
88
82
88
87
95
88
77
78:22 70
77:23 75
85:15 70
90:10 56
78:22 72
83:17 74
2[e]
3
4-MeOC6H4 3c
2-MeOC6H4 3d
2-Naphthyl
PhCH=CH
2-Thienyl
2-Furyl
4
5
3e
3 f
3g
3h
6[f]
7
91:9
86:14 91
98:2 93
84
8
9[f]
Cyclopropyl 3i
[a] Unless otherwise noted, reactions were carried out by addition of sili-
con tetrachloride (1.0 mmol) to a solution of ketone 1 (0.25 mmol), alde-
hyde 2a (0.55 mmol), cHex2NMe (1.25 mmol), and (S)-BINAPO
(10 mol%) in EtCN/CH2Cl2 (1:1, 5 mL) at À608C. [b] Yield of the diaste-
reomers. [c] Determined by 1H NMR spectroscopy (chiro/meso). [d] De-
termined by HPLC analysis. [e] For 48 h. [f] In CH2Cl2 instead of EtCN/
CH2Cl2.
Table 2. Screening of Lewis base catalysts.[a]
but the enantioselectivities decreased slightly (Table 3, en-
tries 3 and 4). The steric effects of the ortho substituent on
the benzene ring on 1d particularly affected the enantiose-
lectivity. Ketone 1e afforded a product with a similar reac-
tivity and selectivity to those observed in acetophenone (1a)
(Table 3, entry 5). Although benzalacetone (1 f), a conjugate
ketone, was less reactive compared to aromatic ketones, the
use of dichloromethane improved the reactivity and gave
the corresponding product in 87% yield (Table 3, entry 6).
Ketones 1g and 1h, which contained heteroaromatic rings,
showed good diastereo- and enantioselectivities (Table 3, en-
tries 7 and 8). The reaction of 2-acetylfuran (1h), in particu-
lar, yielded a high enantioselectivity (91% ee, entry 8). Cy-
clopropyl ketone 1i was less reactive but afforded high dia-
stereo- and enantioselectivities using CH2Cl2 as the solvent
(Table 3, entry 9).
The double aldol reactions of various aldehydes 2 with ke-
tones 1h and 1i were investigated (Table 4). The reaction of
the para-substituted aldehydes 2b and 2c with ketone 1h
produced similar yields and selectivities to those with 2a, al-
though the substituents affected the reactivity of the alde-
hyde (Table 4, entries 2 and 3). Other aromatic aldehydes
also gave the corresponding products in high yields and se-
lectivities (Table 4, entries 4 and 5). Aldehyde 2e, in particu-
lar, provided excellent enantioselectivity (97% ee, Table 4,
entry 5). Cyclopropyl ketone 1i provided the double aldol
product in high yield with high diastereo- and enantioselec-
tivities (Table 4, entries 6–12). The reaction of cinnamalde-
hyde (2 f) gave a lower yield than the reactions of the aro-
matic aldehydes, although good stereoselectivity was ob-
served (Table 4, entry 7). Aromatic aldehydes gave the cor-
responding double aldol adducts in good yields with high di-
astereo- and enantioselectivities (Table 4, entries 8–12).
Entry
Lewis base
Yield
[%][b]
d.r.[c]
ee [%]
AHCTUNGTRENNUNG
(chiro)[d]
1
2
3
4
5
(S)-BINAPO
(S)-tol-BINAPO
(S)-SEGPHOSO
86
88
93
68
21
78:22
76:24
79:21
68:32
69:31
70
70
63
36
ACHTUNGTRENNUNG
À29
[a]All the reactions were carried out by addition of silicon tetrachloride
(1.0 mmol) to solution of ketone 1a (0.25 mmol), aldehyde 2a
(0.55 mmol), cHex2NMe (1.25 mmol), and Lewis base catalyst
a
a
(10 mol%) in EtCN/CH2Cl2 (1:1, 2.5 mL) at À608C. [b] Yield of the dia-
stereomers. [c] Determined by 1H NMR spectroscopy (chiro/meso).
[d] Determined by HPLC analysis.
(S)-SEGPHOS dioxide (SEGPHOSO) showed good reactiv-
ity but the selectivity decreased slightly (Table 2, entry 3).
Lewis bases with electron-donating groups tended to in-
crease the rate of the catalytic cycle. (R,R)-DIOP dioxide
(DIOPO) decreased both the chemical and optical yields
(Table 2, entry 4). The efficiency of another Lewis base cata-
lyst, bipyridine N,N’-dioxide,[14] was tested in the context of
the reaction. However, the catalytic activity was quite low in
comparison with the activities of the phosphine oxides
(Table 2, entry 5).
With the optimal conditions and an organocatalyst in
hand, we explored the scope of the double aldol reaction of
various types of ketones 1 with benzaldehyde (2a) (Table 3).
p-Bromoacetophenone (1b), a ketone with an electron-with-
drawing group, was less reactive and gave the corresponding
adduct in 78% yield after 48 h (Table 3, entry 2).[15] On the
other hand, p- or o-methoxyacetophenone (1c and 1d, re-
spectively), which contained an electron-donating group,
yielded the corresponding adducts in good chemical yields,
Chem. Eur. J. 2011, 17, 7992 – 7995
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7993

S. Kotani, M. Nakajima et al.
Table 4. Scope of aldehyde 2 in the double aldol reaction.[a]
Entry
Ketone
Aldehyde
2
Product
3
Yield
[%][c]
d.r.[d]
ee [%]
ACHTUNGTRENNUNG
(chiro)[e]
1[b]
1
2
3
4
5
6
7
8
1h
1h
1h
1h
1h
1i
1i
1i
1i
1i
2a
2b
2c
2d
2e
2a
2 f
2g
2b
2c
2d
2e
3h
3j
3k
3l
3m
3i
3n
3o
3p
3q
3r
3s
88
90
75
84
80
77
61
76
90
93
96
77
86:14
86:14
83:17
87:13
92:8
91
90
88
85
97
93
85
88
91
90
87
88
98:2
80:20
88:12
97:3
97:3
96:4
9
10
11
12
Scheme 3. Control experiments (reactions (1)–(4)). brsm=based on re-
covered starting material.
1i
1i
95:5
[a] All the reactions were carried out by addition of silicon tetrachloride
(1.0 mmol) to solution of ketone (0.25 mmol), an aldehyde 2
a
1
ty as was produced in the presence of (S)-BINAPO
(Scheme 3, reaction (4)). Therefore phosphine oxide was re-
quired for the second aldol reaction but the chirality of the
catalyst did not influence the stereoselectivity of the second
aldol reaction.
(0.55 mmol), cHex2NMe (1.25 mmol), and (S)-BINAPO (10 mol%) at
À608C. [b] The reaction of 1h was conducted in EtCN/CH2Cl2 (1:1,
5 mL), whereas the reaction of 1i was conducted in CH2Cl2 (5 mL).
[c] Yield of the combined isolated diastereomers. [d] Determined by
1H NMR spectroscopy (chiro/meso). [e] Determined by HPLC analysis.
On the basis of these results, we propose a mechanism for
the double aldol reaction, as shown in Scheme 4. Initially,
the first enantioselective aldol reaction was controlled by
the chiral phosphine oxide to give the optically active tri-
As we have reported previously,[7] the use of a phosphine
oxide is essential to promoting the enantioselective aldol re-
action. In this process, a phosphine oxide catalyzed both the
generation of the trichlorosilyl enol ether and the subse-
quent enantioselective aldol reaction. Further aldolization
of the silylated aldol adduct would have provided the
double aldol product, however, the effect of the phosphine
oxide in the second aldol reaction was unclear. To confirm
this point, the following control experiments were per-
formed (Scheme 3, reactions (1)–(4)). First, the aldol reac-
tion of the racemic aldol adduct 4a in the absence of
BINAPO yielded no products, only the starting material 4a
was recovered (Scheme 3, reaction (1)). This result suggest-
ed that a phosphine oxide was also necessary for the second
aldol transformation. Second, the aldol reaction of the (R)-
or (S)-aldol adduct 4a[16] in the presence of (S)-BINAPO af-
forded the double aldol product 3a in good yield (Scheme 3,
reactions (2) and (3)). Interestingly, despite the use of (S)-
BINAPO in both cases, the absolute configuration of the
starting material 4a was reflected in the configuration of the
product 3a. These results suggested that the enantioselectiv-
ity of the double aldol reaction was determined in the first
aldol process, and that the second aldol process was con-
trolled diastereoselectively by the chirality of the first aldol
product, although small matched/mismatched effects were
observed. Third, the reaction of (R)-4a with racemic
BINAPO produced (R,R)-3a with the same enantioselectivi-
Scheme 4. Proposed reaction mechanism for the double aldol reaction.
chlorosilyl ether 5. Subsequently, intramolecular enolization
of 5 occurred under activation by the phosphine oxide to
provide the chiral cyclic enol ether 6. The enol ether 6 react-
ed diastereoselectively with another aldehyde to give the
chiro product 3 with an enantiomeric excess that reflected
that of the mono-aldol adduct 5.
In conclusion, we have demonstrated, for the first time,
an enantioselective double aldol reaction using silicon tetra-
chloride with a phosphine oxide as an organocatalyst. The
7994
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 7992 – 7995

Enantioselective Double Aldol Reaction
COMMUNICATION
kawa, Chem. Rev. 2002, 102, 2187–2209; c) S. Saito, H. Yamamoto,
Acc. Chem. Res. 2004, 37, 570–579; d) W. Notz, F. Tanaka, C. F. Bar-
bas III, Acc. Chem. Res. 2004, 37, 580–591; e) G. Guillena, C.
Nꢁjera, D. J. Rꢂamn, Tetrahedron: Asymmetry 2007, 18, 2249–2293;
f) S. Mukherjee, J. W. Yang, S. Hoffmann, B. List, Chem. Rev. 2007,
107, 5471–5569.
cyclic enolization of the first aldol product enabled the sub-
sequent aldol reaction to proceed with good yields and ste-
reoselectivities. Several control experiments suggested that
the enantio-determining step may have been the first aldol
reaction and that the second step proceeded by the diaste-
reoselective enolization of the mono-aldol adduct. Further
investigations to improve the diastereo- and enantioselectiv-
ities and to evaluate the detailed mechanism are currently
in progress.
[6] For leading and recent papers on direct asymmetric aldol reactions,
see: a) Y. M. A. Yamada, N. Yoshikawa, H. Sasai, M. Shibasaki,
Angew. Chem. 1997, 109, 1942–1944; Angew. Chem. Int. Ed. Engl.
1997, 36, 1871–1873; b) B. List, R. A. Lerner, C. F. Barbas III, J.
Am. Chem. Soc. 2000, 122, 2395–2396; c) B. M. Trost, H. Ito, J. Am.
Chem. Soc. 2000, 122, 12003–12004; d) T. Suzuki, N. Yamagiwa, Y.
Matsuo, S. Sakamoto, K. Yamaguchi, M. Shibasaki, R. Noyori, Tet-
rahedron Lett. 2001, 42, 4669–4671; e) S. S. V. Ramasastry, K. Al-
bertshofer, N. Utsumi, C. F. Barbas III, Org. Lett. 2008, 10, 1621–
1624; f) Y. Hayashi, T. Itoh, S. Aratake, H. Ishikawa, Angew. Chem.
2008, 120, 2112–2114; Angew. Chem. Int. Ed. 2008, 47, 2082–2084;
g) K. Nakayama, K. Maruoka, J. Am. Chem. Soc. 2008, 130, 17666–
17667; h) K. Baer, M. Kraußer, E. Burda, W. Hummel, A. Berkessel,
H. Grçger, Angew. Chem. 2009, 121, 9519–9522; Angew. Chem. Int.
Ed. 2009, 48, 9355–9358; i) M. Iwata, R. Yazaki, Y. Suzuki, N. Ku-
magai, M. Shibasaki, J. Am. Chem. Soc. 2009, 131, 18244–18245;
j) H. Ube, N. Shimada, M. Terada, Angew. Chem. 2010, 122, 1902–
1905; Angew. Chem. Int. Ed. 2010, 49, 1858–1861.
Experimental Section
Representative procedure: Silicon tetrachloride (0.11 mL, 1.0 mmol,
4.0 equiv) was added to a solution of aldehyde 2a (0.056 mL, 0.55 mmol,
2.2 equiv), ketone 1a (0.027 mL, 0.25 mmol, 1.0 equiv), dicyclohexylme-
thylamine (0.27 mL, 1.25 mmol, 5.0 equiv), and (S)-BINAPO (16.4 mg,
0.025 mmol, 10 mol%) in CH2Cl2 (1.25 mL) and propionitrile (1.25 mL)
at À608C. After being stirred for 24 h, the reaction was quenched with
sat. NaHCO3 (3 mL) and then the slurry was stirred for 0.5 h. The two-
layer mixture was filtered through a celite pad and the aqueous layer was
extracted with EtOAc (40 mL). The combined organic layers were
washed with 10% HCl (20 mL), sat. NaHCO3 (20 mL), and brine
(20 mL) then dried over Na2SO4. After filtration and concentration, the
obtained crude product was purified by column chromatography
(hexane/EtOAc=4:1, SiO2 10 g) to give meso-3a (15.7 mg) and chiro-3a
(55.8 mg) (total yield: 86%, d.r.=78:22). The enantiomeric excess was
determined to be 70% ee (chiro) by HPLC analysis with Daicel Chiral-
pak AD-H column (eluent: 9:1 hex/IPA; flow rate: 1.0 mLminÀ1; detec-
tion: 254 nm; Rt: 17.0 min for (S,S)-isomer, 22.3 for (R,R)-isomer).
[7] a) S. Kotani, Y. Shimoda, M. Sugiura, M. Nakajima, Tetrahedron
Lett. 2009, 50, 4602–4605; b) S. Kotani, S. Aoki, M. Sugiura, M. Na-
kajima, Tetrahedron Lett. 2011, 52, 2834–2836.
[8] We have also demonstrated that chiral phosphine oxides catalyze
the aldol reaction of trichlorosilyl enol ethers to provide high dia-
stereo- and enantioselectivities: a) S. Kotani, S. Hashimoto, M. Na-
kajima, Synlett 2006, 1116–1118; b) S. Kotani, S. Hashimoto, M. Na-
kajima, Tetrahedron 2007, 63, 3122–3132; for related chiral phos-
phine oxide-catalyzed reductive aldol reactions, see: c) M. Sugiura,
N. Sato, S. Kotani, M. Nakajima, Chem. Commun. 2008, 4309–4311;
d) M. Sugiura, N. Sato, Y. Sonoda, S. Kotani, M. Nakajima, Chem.
Asian J. 2010, 5, 478–481.
Acknowledgements
[9] For reviews on Lewis base organocatalysts, see: a) Y. Orito, M. Na-
ˇ
´
kajima, Synthesis 2006, 1391–1401; b) A. V. Malkov, P. Kocovsky,
Eur. J. Org. Chem. 2007, 29–36; c) S. E. Denmark, G. L. Beutner,
Angew. Chem. 2008, 120, 1584–1663; Angew. Chem. Int. Ed. 2008,
47, 1560–1638; d) M. Benaglia, S. Rossi, Org. Biomol. Chem. 2010,
8, 3824–3830.
This work was partially supported by a Grant-in-Aid for Scientific Re-
search from the Ministry of Education, Culture, Sports, Science and
Technology of Japan.
[10] The diastereomers were represented as chiro and meso isomers ac-
cording to the nomenclature of the isomers of inositol.
[11] The chiro- and meso-3a are separable by silica gel chromatography.
The enantiomeric excess of chiro isomer can be determined by
HPLC analysis even as a diastereomeric mixture (see the Support-
ing Information for details).
[12] Other solvents such as tetrahydrofuran or toluene did not yield the
double aldol products.
[13] The high nucleophilicities and/or low basicities of amines such as
triethylamine and pyridine might interrupt the production of the
double aldol adduct 3a.
[14] (R)-BQNO was prepared according to the literature, see: M. Nakaji-
ma, M. Saito, M. Shiro, S. Hashimoto, J. Am. Chem. Soc. 1998, 120,
6419–6420.
[15] Small amounts (<5%) of mono-aldol adduct were obtained in most
cases.
[16] (R)- and (S)-4a were prepared following the procedure, see: M. Na-
kajima, Y. Orito, T. Ishizuka, S. Hashimoto, Org. Lett. 2004, 6,
3763–3765.
Keywords: aldol reaction · asymmetric catalysis · domino
reactions · intramolecular enolization · Lewis base
[1] For reviews of sequential reactions, see: a) L. F. Tietze, U. Beifuss,
Angew. Chem. 1993, 105, 137–170; Angew. Chem. Int. Ed. Engl.
1993, 32, 131–163; b) L. F. Tietze, Chem. Rev. 1996, 96, 115–136.
[2] For borane-mediated double aldol reactions, see: a) A. Abiko, J.-F.
Liu, D. C. Buske, S. Moriyama, S. Masamune, J. Am. Chem. Soc.
1999, 121, 7168–7169; b) A. Abiko, T. Inoue, S. Masamune, J. Am.
Chem. Soc. 2002, 124, 10759–10764; c) H. Furuno, T. Inoue, A.
Abiko, Tetrahedron Lett. 2002, 43, 8297–8299; d) A. Abiko, Acc.
Chem. Res. 2004, 37, 387–395.
[3] For a triple aldol cascade reaction, see: B. J. Albert, H. Yamamoto,
Angew. Chem. 2010, 122, 2807–2809; Angew. Chem. Int. Ed. 2010,
49, 2747–2749.
[4] For stepwise double aldol reactions, see: a) T. Mukaiyama, H. Arai,
I. Shiina, Chem. Lett. 2000, 580–581; b) H. Arai, I. Shiina, T. Mu-
kaiyama, Chem. Lett. 2001, 118–119; c) T. Mukaiyama, K. Pubhom,
K. Yamane, H. Arai, Bull. Chem. Soc. Jpn. 2003, 76, 413–425.
[5] For reviews on direct aldol reactions, see: a) B. Alcaide, P. Almen-
dros, Eur. J. Org. Chem. 2002, 1595–1601; b) M. Shibasaki, N. Yosh-
Received: March 25, 2011
Published online: May 30, 2011
Chem. Eur. J. 2011, 17, 7992 – 7995
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7995