Quinox, a quinoline-type N-oxide, as organocatalyst in the asymmetric allylation of aromatic aldehydes with allyltrichlorosilanes: The role of arene-arene interactions

Article abstract of DOI:10.1002/anie.200351737

The allylation of aromatic aldehydes with allyltrichlorosilanes can be catalyzed by the new Lewis basic organocatalyst quinox with high enantioselectivity for electron-poor aldehydes and low for their electron-rich congeners (see scheme). This behavior su

Full text of DOI:10.1002/anie.200351737

Communications
Asymmetric Allylation
ative 6 and its o-methoxy analogue 7 (each of 87%
enantiopurity) induced the formation of the allylation prod-
uct 3 in 41% and 69% ee, respectively (3, Ar= Ph).[6]
The latter behavior seems to suggest that arene–arene
interactions (e.g., p-stacking)[7] of the incoming benzaldehyde
and the second aromatic nucleus may be involved in the
reaction.[8] Furthermore, the N-oxide catalysts are only
effective with aromatic and heteroaromatic aldehydes,[3–6]
which appears to lend additional credence to this ration-
alization. Hence, if the catalyst contains an electron-rich
aromatic system, such as the o-methoxy-phenyl in 7, it should
be most effective with electron-poor aromatic aldehydes (i.e.,
with electron-withdrawing substituent) and vice versa. In line
with this hypothesis, we have endeavored to synthesize a new
electron-rich catalyst and explore its activity toward electron-
rich and electron-poor aromatic aldehydes. Moreover, ben-
zaldehyde derivatives containing electron-withdrawing
groups were reported to produce only modest enantioselec-
tivities in the allylation reaction,[3–6] so that development of an
effective catalyst for this class of compounds would be
desirable.
Quinox, a Quinoline-Type N-Oxide, as
Organocatalyst in the Asymmetric Allylation of
Aromatic Aldehydes with Allyltrichlorosilanes:
The Role of Arene–Arene Interactions**
Andrei V. Malkov,* Lenka Dufkovµ, Louis Farrugia,
ˇ
´
and Pavel Kocovsky*
Dedicated to Professor Rudolf Zahradník
on the occasion of his 75th birthday
Activation of allylsilane reagents 2 by a catalytic amount of a
chiral Lewis base has been shown to effect asymmetric
allylation of aromatic and heteroaromatic aldehydes 1
(Scheme 1).[1–4] The highest enantioselectivities were reported
We have chosen the isoquinoline N-oxide derivative 11 as
the candidate catalyst (Scheme 2), the synthesis of which was
inspired by the simplicity of the initial steps toward the well-
known heterobidentate ligand quinap.[9] Thus, the Suzuki–-
Miyaura coupling of 1-chloro-isoquinoline (8)[9b] with boronic
acid 9[9b,10] afforded the biaryl derivative 10[9] (95%), whose
treatment with m-chloroperoxybenzoic acid provided racemic
N-oxide (Æ )-11 (99%).
Scheme 1. Asymmetric allylation of aromatic and heteroaromatic alde-
hydes with a Lewis base catalyst.
by us for pindox 4 (ꢀ 92% ee) and its dimethyl analogue 5
(ꢀ 98% ee) as catalysts (3, Ar= Ph).[5] We also proposed a
mechanism consistent with the available data, in which we
postulated the chelation of the silicon atom by the oxygen
atom of the N-oxide and the nitrogen atom of the second
pyridine ring of 4/5. However, in a subsequent study, we
demonstrated that the second nitrogen is not a prerequisite
for good asymmetric induction, as the simple phenyl deriv-
ˇ
´
[*] Dr. A. V. Malkov, Prof. Dr. P. Kocovsky, L. Dufkovµ, Dr. L. Farrugia
Department of Chemistry
Joseph Black Building
University of Glasgow
Glasgow G12 8QQ (UK)
Scheme 2. Synthesis of the isoquinoline N-oxide derivative 11.
Fax: (+44)141-33-488
E-mail: amalkov@chem.gla.ac.uk
p.kocovsky@chem.gla.ac.uk
Racemic 11 was resolved by cocrystallization with (S)-
(À)-binol (12; binol = 2,2'-dihydroxy-1,1'-biphenyl),[11] which
gave the crystalline material containing binol and (+)-11 (in a
1:1 ratio), while (À)-11 remained in the solution. This
cocrystallization, followed by a chromatographic separation
of (+)-11 from (S)-binol, furnished pure (+)-11 of 98% ee (as
revealed by chiral HPLC) in 89% yield of the isolated
product.[12] The absolute configuration of 11 was found to be
(R)-(+)-11 by crystallographic analysis of the molecular
L. Dufkovµ
Department of Organic Chemistry
Charles University
128 40 Prague 2 (Czech Republic)
[**] We thank the University of Glasgow, the Socrates Exchange
program, and Dr. Alfred Bader for financial support.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3674
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200351737
Angew. Chem. Int. Ed. 2003, 42, 3674 –3677

Angewandte
Chemie
crystal of (+)-11 with (S)-(À)-12 (Figure 1),[13] the absolute
configuration of which is known. In line with the accepted
acronym quinap,[9] we propose the new acronym quinox for 11.
reaction was slower and rather less enantioselective (entry 3).
A further decrease of the enantioselectivity was observed for
the reaction in CHCl3 (63% ee, entry 4) but the reaction was
much faster (30 min at À408C and ~ 1 h at À608C; Table 1,
entries 4 and 5). Cinnamyl derivatives 1b and 1c had good
reactivity with modest enantioselectivity (entries 6 and 7).
Electron-rich p-methoxybenzaldehyde (1d) and the 2-
furyl and 2-thiophenyl analogues 1e and 1 f gave almost
racemic products in good yields (Table 1, entries 8–10), while
pyridyl aldehydes 1 g and 1h proved to be practically
unreactive (entries 11 and 12).
By contrast, the introduction of electron-withdrawing
substituents into p-position resulted in a dramatic increase in
both reactivity and enantioselectivity. Thus, p-nitrobenzalde-
hyde 1i afforded the corresponding product in better yield
and with slightly higher ee than benzaldehyde (compare
entries 1 and 13), and the p-halo derivatives 1j and 1k both
gave > 90% ee (Table 1, entries 14 and 15). The highest
conversion and enantioselectivity (96% ee) was attained with
the p-trifluoromethyl derivative 1 l (entry 16).
The observed trend in the reaction rate and enantiose-
lectivity, with best results obtained for the electron-poor
benzaldehydes 1i–l, is fully compatible with the original
hypothesis of the arene–arene interaction of the catalyst with
the incoming aldehyde. The acceleration in chloroform
appears to lend further support to these interactions as the
driving force for the reaction.[15]
Figure 1. An ORTEP diagram illustrating the interaction of (R)-(+)-11
(on the left) with (S)-(À)-12 (right), in particular the hydrogen bonding
À
À
N O···H O. Displacement parameters are shown at the 50% probabil-
ity level.
The addition of allyltrichlorosilane (2) to benzaldehyde
(1a) (Scheme 1), carried out in the presence of (R)-(+)-11
(5 mol%) at À408C for 2 h in CH2Cl2, produced (R)-(+)-3a
The allylation with trans-crotyltrichlorosilane 13 (pre-
of 87% ee (Table 1, entry 1). When the catalyst load was pared as an 87:13 trans/cis mixture by the CuCl-catalyzed
lowered to 1 mol% the reaction slowed (to ~ 12 h) but the
reaction of crotyl chloride with HSiCl3)[2b] was briefly
explored to assess the scope of the reaction and to shed
more light on the mechanism (Scheme 3). With pindox (4) as
catalyst, the reaction in CH2Cl2 produced mainly anti-14
(Table 2, entry 1), which is compatible with the generally
accepted cyclic transition state A.[1,3] By contrast, a 2:1 anti/
syn mixture (entry 2) was produced in the presence of quinox
(11), suggesting a participation of the open-chain transition
state B. In MeCN, the enantioselectivity was significantly
reduced but the anti/syn ratio was increased (entry 3). A
further drop in enantioselectivity and simultaneous increase
of diastereoselectivity was observed for the electron-rich
aldehyde 1d (entry 4). By contrast, the electron-poor alde-
hyde 1l (entries 5 and 6), whose arene–arene interactions
with the catalyst are assumed to be stronger, exhibited high
enantioselectivity and lower diastereoselectivity. This trend
suggests that, in the case of quinox (11), the cyclic transition
enantioselectivity was not altered (entry 2).[14] In MeCN, the
Table 1: The Allylation of Aldehydes 1 with 2 Catalyzed by (R)-(+)-11
(Scheme 1).[a]
entry aldehyde Ar
solvent t [h] yield [%][b] ee [%][c,d]
1
2
3
4
5
6
7
1a
1a
1a
1a
1a
1b
1c
Ph
Ph
Ph
Ph
CH2Cl2
CH2Cl2 12
2
60
55
60
87
87
70
63
62
51
55
[e]
MeCN
CHCl3
12
0.5 80
1
[f]
Ph
CHCl3
79
86
71
Cinnamyl
a-Methylcin-
namyl
CH2Cl2 12
CH2Cl2 12
8
9
1d
1e
1 f
1g
1h
1i
1j
1k
1l
4-MeO-C6H4
2-Furyl
CH2Cl2 12
CH2Cl2 12
CH2Cl2 12
CH2Cl2 12
CH2Cl2 12
70
68
59
12
5
6
–
–
89
93
91
96
10
11
12
13
14
15
16
2-Thiophenyl
2-Pyridyl
4-Pyridyl
4-NO2-C6H4
4-Cl-C6H4
4-F-C6H4
4-CF3-C6H4
25[g]
trace
73
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
2
2
2
2
65
79
85
[a] The reaction was carried out at 0.4 mmol scale with 1.1 equiv of 2, in
the presence of (R)-(+)-11 (5 mol%, 98% ee) as catalyst and (i-Pr)2NEt
(1 equiv) as base at À408C. [b] Yield of the isolated product (note that
some of the products are fairly volatile). [c] Determined by chiral HPLC or
GC. [d] All products 3 were of (R)-(+)-configuration, as revealed by the
comparison of their optical rotations (measured in CHCl3) and their GC
and HPLC retention times with the literature data and with the behavior
of authentic samples.[3a,b,16,17] [e] With 1 mol% of the catalyst. [f] At
À608C. [g] For product identification, see reference [18,19].
Scheme 3. Proposed reaction transition states, cat.*=chiral catalyst.
Angew. Chem. Int. Ed. 2003, 42, 3674 –3677
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3675

Communications
Table 2: The Allylation of Aldehydes 1 with 13 (Scheme 3).[a]
thane (20 mL) and the resulting clear
solution was allowed to cool to room
temperature. The molecular complex
(R)-11·(S)-12 thus formed as a white
precipitate over the period of about
30 min was collected by suction filtra-
tion and the individual components
were separated by column chromatog-
raphy on silica gel (1 ” 20 cm) with
dichloromethane, which eluted (S)-
(À)-12 (235 mg, 49%), followed by a
entry
aldehyde
Ar
catalyst
solvent
yield [%][b]
anti:syn
ee [%][c,d] anti, syn
1
2
3
4
5
6
1a
1a
1a
1d
1l
Ph
Ph
Ph
(+)-4[e]
CH2Cl2
CH2Cl2
MeCN
MeCN
CH2Cl2
MeCN
54
70
51
53
76
62
93:7
87[g]
(+)-11[f]
(+)-11[f]
(+)-11[f]
(+)-11[f]
(+)-11[f]
68:32
76:24
82:18
70:30
60:40
65, 78
56, 60
50, 37
92, 95
80, 84
4-MeO-C6H4
4-CF3-C6H4
4-CF3-C6H4
1l
[a] The reaction was carried out in the same way as shown in Table 1. at À408C. [b] Isolated yield (note
that some of the products are fairly volatile). [c] Determined by chiral HPLC or GC. [d] The products 14
had the absolute configuration shown in Scheme 3, as revealed by the comparison of their optical
rotations (measured in CHCl3) and their GC retention times with the literature data[3a] and with the
behavior of authentic samples. [e] At À608C for 24 h. [f] At À408C for 12 h. [g] Enantiomer of anti-14 was
formed with (+)-4 as catalyst.
dichloromethane/methanol
mixture
(93:7), which eluted (R)-(+)-11
(223 mg, 45%, or 89% when calculated
for a single enantiomer): mp 178–1818C
(ethyl acetate-hexane); [a]2D0 = + 134
(c = 1.33 in CHCl3) (lit. [3d] gives mp
180.58C and [a]D = + 132.1 (c = 0.9, CHCl3); chiral HPLC (Chiralcel
AD-H, hexane/2-propanol 75:25, 1 mLminÀ1) showed 96–98% ee,
depending on the batch (tS = 12.36 min, tR = 19.17 min). The filtrate
from the original crystallization was evaporated and the components
of the residue were separated by chromatography in the same manner
as that shown for the molecular crystal to furnish (S)-(À)-12 (224 mg,
47%), followed by (S)-(À)-11 (215 mg, 43%, or 86% calculated for a
single enantiomer), which was of 70% ee.
General procedure for the asymmetric allylation of aldehydes 1
with allyltrichlorosilane (2): Allyltrichlorosilane (75 mL, 0.47 mmol)
was added to a solution of catalyst (R)-(+)-11 (6 mg, 0.02 mmol or
1.2 mg, 0.004 mmol), diisopropylethylamine (46 mL, 0.5 mmol), and
aldehyde (0.4 mmol) in dichloromethane (2 mL) under nitrogen at
À408C. The reaction mixture was stirred at À408C for 0.5–12 h (see
Table 1). The reaction was quenched with saturated aqueous
NaHCO3 (1 mL), the layers were separated, and the aqueous layer
was extracted with dichloromethane (2 ” 5 mL). The combined
organic extracts were washed with saturated aqueous NaCl (3 mL)
and dried over Na2SO4. The solvent was removed in vacuo and the
residue was purified by column chromatography on silica gel (1 ”
20 cm) with a petroleum ether/ethyl acetate mixture (9:1). The yields
and enantioselectivities are given in Table 1.
state A is less favored when arene–arene interactions operate
(Table 2, entries 2 and 5), and more favored when the latter
interactions are minimal (entry 4). Note that 1d has the more
Lewis basic carbonyl oxygen and should be more prone to
coordinate to the Lewis acidic silicon, which should favor A.
On the other hand, the less Lewis basic 1a and 1l are less
suitable for this coordination, leaving B as an option, which is
compatible with the experimental results.[20]
In conclusion, quinox (R)-(+)-11 has been synthesized
and shown to exercise an unusually high level of enantiocon-
trol in the Sakurai–Hosomi–Denmark-type allylation of
electron-poor aromatic aldehydes (ꢀ 96% ee, the highest
value reported to date). An arene–arene interaction between
the catalyst and the substrate aldehyde has been proposed as
a rationale for this observation.[21] These reactions require low
catalyst loading (ꢀ 5 mol%) and are characterized by a
substantial solvent effect upon the rate (typically 12 h in
MeCN, 2 h in CH2Cl2, and 30 min in CHCl3).
Experimental procedures, analytical and spectral data, crystallo-
graphic data, and copies of the NMR spectra for the key compounds
are available in the Supporting Information.
Experimental Section
( Æ )-1-(2-Methoxy-1-naphthyl)-isoquinoline-N-oxide (Æ )-11: m-
Chloroperoxybenzoic acid (70%, 2.2 g, 9.3 mmol) was added to a
solution of 1-(2-methoxy-1-naphthyl)isoquinoline 10[9b] (1.32 g,
4.6 mmol) in dichloromethane (20 mL) at 08C and the mixture was
stirred at room temperature for 2 h. The mixture was then extracted
with saturated NaHCO3 (10 mL), the organic solution was washed
with brine (5 mL), dried over Na2SO4, and evaporated in vacuo. The
crude product was purified by column chromatography on silica gel
(1 ” 20 cm) with an ethyl acetate/methanol mixture (4:1) to afford
(Æ )-11 as white solid (1.37 g, 98%): mp 107–1108C (ethyl acetate/
hexane); 1H NMR (400 MHz, CDCl3, 258C): d = 3.75 (s, 3H), 6.96 (d,
J = 8.0 Hz, 1H), 7.06 (d, J = 8.0 Hz, 1H), 7.20–7.32 (m, 3H), 7.37 (d,
J = 8.8 Hz, 1H), 7.43–7.47 (m, 1H), 7.67 (d, J = 6.8 Hz, 1H), 7.76–7.82
(m, 2H), 8.10 (d, J = 8.8 Hz, 1H), 8.39 ppm (d, J = 7.2 Hz, 1H);
13C NMR (100 MHz, CDCl3, 258C): d = 57.03 (CH3), 113.91 (CH),
113.96 (C), 123.96 (CH), 124.23 (CH), 124.42 (CH), 125.88 (CH),
127.27 (CH), 127.90 (CH), 128.64 (CH), 128.80 (CH), 129.20 (C),
129.53 (CH), 129.57 (CH), 130.73 (C), 132.23 (CH), 132.94 (C), 138.03
(CH), 143.84 (C), 155.91 ppm (C); IR (KBr): n˜ = 3054 (w), 1319 (s),
1268 (s), 1251 cmÀ1 (s); MS (EI, 70 eV) m/z (%) 301.4 (35) [MC+],
284.4 (87), 270.3 (100), 242.3 (47), 241.3 (40), 120.7 (19); HRMS (EI)
301.1102 (C20H15O2N requires 301.1103).
Received: April 24, 2003 [Z51737]
Keywords: allylation · arenes · asymmetric catalysis · O ligands ·
.
organocatalysis
[1] a) S. E. Denmark, D. M. Coe, N. E. Pratt, B. D. Griedel, J. Org.
Chem. 1994, 59, 6161; b) S. E. Denmark, X. Su, Y. Nishigaichi, J.
Am. Chem. Soc. 1998, 120, 12990; c) S. E. Denmark, R. A.
Stavenger, K.-T. Wong, X. Su,J. Am. Chem. Soc. 1999, 121, 4982;
d) S. E. Denmark, R. A. Stavenger, K.-T. Wong, X. Su, J. Am.
Chem. Soc. 1999, 121, 4982; e) S. E. Denmark, J. Fu, J. Am.
Chem. Soc. 2000, 122, 12021; f) S. E. Denmark, S. M. Pham,
Helv. Chim. Acta 2000, 83, 1846; g) S. E. Denmark, J. Fu, J. Am.
Chem. Soc. 2001, 123, 9488; h) S. E. Denmark, Y. Fan, J. Am.
Chem. Soc. 2002, 124, 4233; i) S. E. Denmark, J. Fu, J. Am. Chem.
Soc. 2003, 125, 2208. For overviews, see: j) S. E. Denmark, R. A.
Stavenger, Acc. Chem. Res. 2000, 33, 432; k) S. E. Denmark, J.
Fu, Chem. Commun. 2003, 167.
[2] a) K. Iseki, Y. Kuroki, M. Takahashi, Y. Kobayashi, Tetrahedron
Lett. 1996, 37, 5149; b) K. Iseki, Y. Kuroki, M. Takahashi, S.
Kishimoto, Y. Kobayashi, Tetrahedron 1997, 53, 3513.
[3] a) M. Nakajima, M. Saito, M. Shiro, S. Hashimoto, J. Am. Chem.
Soc. 1998, 120, 6419; b) M. Nakajima, M. Saito, S. Hashimoto,
(R)-(+)-1-(2-Methoxy-1-naphthyl)-isoquinoline-N-oxide
(+)-11: Solid (S)-(À)-binol (S)-(À)-12 (475 mg, 1.66 mmol) was
added to hot solution of racemic 1-(2-methoxy-1-naphthyl)-
(R)-
a
isoquinoline-N-oxide (Æ )-11 (500 mg, 1.66 mmol) in dichlorome-
3676
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2003, 42, 3674 –3677

Angewandte
Chemie
Chem. Pharm. Bull. 2000, 48, 306; c) M. Nakajima, M. Saito, S.
Hashimoto, Tetrahedron: Asymmetry 2002, 13, 2449; d) M.
Nakajima, M. Saito, M. Uemura, S. Hashimoto, Tetrahedron
Lett. 2002, 43, 8827.
[4] T. Shimada, A. Kina, S. Ikeda, T. Hayashi, Org. Lett. 2002, 4,
2799.
[5] A. V. Malkov, M. Orsini, D. Pernazza, K. W. Muir, V. Langer, P.
ˇ
´
Meghani, P. Kocovsky, Org. Lett. 2002, 4, 1047.
ˇ
´
[6] A. V. Malkov, M. Bell, M. Vassieu, V. Bugatti, P. Kocovsky, J.
Mol. Catal. A 2003, 196, 179.
[7] For a recent account on the face-to-face and center-to-edge
interactions on aromatic systems and polar effects, see: a) F.
Cozzi, R. Annunziata, M. Benaglia, M. Cinquini, L. Raimondi,
K. K. Baldridge, J. S. Siegel, Org. Biomol. Chem. 2003, 1, 157;
b) For a recent review on the arene–arene, p stacking, and C-H/
p interactions, see: E. A. Meyer, R. K. Castellano, F. Diederich,
Angew. Chem. 2003, 115, 1244; Angew. Chem. Int. Ed. 2003, 42,
1210; c) For C-H/p interactions, see: P. Hobza, Z. Havlas, Chem.
Rev. 2000, 100, 4253; d) For a review on p shielding in organic
synthesis, see: G. B. Jones, Tetrahedron 2001, 57, 7999.
[8] The p stacking has also been proposed by Hayashi to account for
the high catalytic activity of
a chiral 2,2’-bipyridine-N,N-
bisoxide.[4]
[9] a) J. M. Brown, D. I. Hulmes, T. P. Layzell, J. Chem. Soc. Chem.
Commun. 1993, 1673; b) N. W. Alcock, J. M. Brown, D. I.
Hulmes, Tetrahedron: Asymmetry 1993, 4, 743.
ˇ
ˇ
ˇ
ˇ
ˇ
[10] S. Vyskocil, L. Meca, I. Tislerovꢀ, I. Cꢁsarovꢀ, M. Polꢀsek, S. R.
Harutyunyan, Y. N. Belokon, R. M. J. Stead, L. Farrugia, S. C.
ˇ
´
Lockhart, W. L. Mitchell, P. Kocovsky, Chem. Eur. J. 2002, 8,
4633.
[11] For the previous use of binol to resolve bipyridine-type N,N-
bisoxides, see refs [3,4].
[12] While this work was in progress, Nakajima published the same
synthesis of (+)-11, including the resolution with (S)-(À)-
binol.[3d] However, this synthesis was mentioned as a footnote
without specifying the conditions and the absolute configuration
of (+)-11 was not determined.
[13] Crystal data for (R)-11·(S)-12: colorless crystals, space group
P212121, a = 8.5583(1) ꢂ, b = 11.9097(1) ꢂ, c = 29.0481(4) ꢂ, V=
2960.78(6) ꢂ3, Z = 4,
RF(obs) = 0.0395.
d
calcd = 1.318 gcmÀ3
,
m = 0.085 mmÀ1
,
[14] Investigation of the allylation of 1a with 2, catalyzed by (R)-(+)-
11 of 29%, 50%, and 75% ee, respectively, has demonstrated a
fully linear relationship between the enantiopurity of the catalyst
and the product. In these experiments, the resulting product 3a
was of 27%, 45%, and 71% ee, respectively.
[15] Chloroform has been shown to be the solvent that most strongly
stabilizes the arene–arene interactions: G. A. Breault, C. A.
Hunter, P. C. Mayers, J. Am. Chem. Soc. 1998, 120, 3402.
[16] Y. Yamamoto, H. Yatagai, K. Maruyama, J. Am. Chem. Soc.
1981, 103, 1969.
[17] H. Doucet, M. Santelli, Tetrahedron: Asymmetry 2000, 11, 4163.
[18] U. S. Racherla, Y. Liao, H. C. Brown, J. Org. Chem. 1992, 57,
6614.
[19] J. Uenishi, T. Hiraoka, S. Hata, K. Nishiwaki, O. Yonemitsu, K.
Nakamura, H. Tsukube, J. Org. Chem. 1998, 63, 2481.
[20] For a discussion on cyclic/open transition state, see: a) S. E.
Denmark, N. G. Almstead, J. Org. Chem. 1994, 59, 5130; b) S. E.
Denmark, S. Hosoi, J. Org. Chem. 1994, 59, 5133.
[21] Hayashi has proposed p–p stacking of 1 and the catalysts but, in
his case, the variation of the enantioselectivity was less dramatic
(94% ee for 1d and 56% ee for 1l as the extremes of the scale).[4]
Furthermore, his catalyst gave the best results in MeCN, which is
known not to support arene–arene interactions,[15] thus suggest-
ing that his and our catalyst may operate through a different
mode of interactions.
Angew. Chem. Int. Ed. 2003, 42, 3674 –3677
www.angewandte.org
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3677