Angewandte
Chemie
ucts.[14] Therefore, we turned our attention to developing a
new catalyst suitable for b-adrenoceptor agonists syntheses.
We recently reported the utility of dinucleating Schiff base
1-H4 (Scheme 1) in nitro-Mannich reactions of N-Boc imines
(Boc = tert-butylcarboxy) with nitroethane and nitropro-
pane.[15] Schiff base 1-H4 selectively incorporated Cu into
the inner N2O2 cavity and an oxophilic rare earth metal,
having a large ionic radius, into the outer O4 cavity. The
cooperative functions of the two metals[16,17] in the hetero-
bimetallic Cu/Sm/1 complex (Scheme 1; M = Cu, RE = Sm)
were key to achieving high diastereo- and enantioselectivity
in the nitro-Mannich reactions. An achiral 4-tert-butylphenol
additive improved the enantioselectivity by performing as an
achiral ligand. We hypothesized that suitable selection of a
dinucleating Schiff base, a transition metal (M)/rare earth
metal (RE) combination, and a phenolic additive would
afford an optimal chiral environment for the anti-selective
nitroaldol reaction. Thus, we initiated optimization reactions
by using Schiff base 1-H4, a phenolic additive, aldehyde 3a,
and nitroethane 4a (Table 1). The Cu/Sm/1 and 4-tert-
butylphenol system, which was optimal for the nitro-Mannich
reactions, gave poor reactivity and selectivity (Table 1,
entry 1). Screening of other rare earth metals (Table 1,
entries 1-4) indicated that La(O-iPr)3 had the best reactivity
(Table 1, entry 4), and additional optimization with regard to
the inner metal (Table 1, entries 4–7) revealed that the best
combination was Pd(OAc)2 and La(O-iPr)3. These conditions
gave 5aa in 82% yield, anti/syn = 5.3:1, and 58% ee (Table 1,
entry 7). Other metals such as Ni(OAc)2 and Zn(OAc)2 gave
less satisfactory results (Table 1, entries 5–6). The phenolic
additive also affected both the diastereo- and enantioselec-
tivity (Table 1, entries 7-9), and 4-bromophenol was found to
be optimal (Table 1, entry 9). Finally, minor modifications of
the solvent and the reaction time
Table 1: Optimization of the reaction conditions.
Entry M[a] RE[b] ArOHSolvent
Yield
d.r.
[%] anti/
syn[c]
ee
[%][f]
[d]
[d]
1
2
3
4
5
6
7
8
Cu Sm 4-tBuC6H4OHTFH
Cu Gd 4-tBuC6H4OHTFH
Cu Dy 4-tBuC6H4OHTFH
33
26
25
73
61
30
82
2.3:1
2.3:1
2.8:1
2:1
2:1
1:2
1
4
3
28
12
2
Cu La
Ni La
Zn La
Pd La
Pd La
4-tBuC6H4OHTFH
4-tBuC6H4OHTFH
4-tBuC6H4OHTFH
4-tBuC6H4OHTFH
4-MeO-
5.3:1
58
THF
65
3.3:1
49
84
C6H4OH
9
Pd La
4-BrC6H4OHFTH
4-BrC6H4OHFT/H
77
xylenes
12:1
92
77
19:1
10[e] Pd La
[a] M(OAc)2 was used. [b] RE(O-iPr)3 was used. [c] Determined by
1HNMR analysis. [d] ent-5aa was the major product. [e] Reaction time
was 69 h. [f] Values determined for the anti product.
92–77% ee, albeit with modest anti selectivity (Table 2,
entries 10–12). The reaction with nitropropane (4b) as a
donor proceeded smoothly to give product 5ab in anti/syn =
19:1 and 85% ee (Table 2, entry 13). By using a 5 mol%
catalyst loading, good anti selectivity and enantioselectivity
were maintained, but a long reaction time was required
(Table 2, entry 14).
4-benzyloxybenzaldehyde 3k was selected for the syn-
thesis of 2a and 2b. A catalytic asymmetric nitroaldol
gave the optimum results in THF/
Table 2: anti-Selective nitroaldol reactions with various aldehydes and nitroalkanes.[a]
xylenes, producing 5aa in 92%
yield, anti/syn = 19:1, and 84% ee
(Table 1, entry 10).
The substrate scope and limita-
tions are shown in Table 2. Aro-
matic aldehydes with electron-
donating substituents at the para-,
meta-, or ortho-position gave prod-
ucts with high anti selectivity and
good enantioselectivity (Table 2,
entries 2–7). For the less reactive
aldehyde 3e, the reaction at À308C
was required for good conversion
(Table 2, entry 6 versus entry 7),
and aldehyde 3 f having an elec-
Entry
R
3
R’
4
Product
t
[h]
Yield[b]
[%]
d.r.
ee
anti/syn[c]
[%][f]
1
2
C6H5
3a
3b
3b
3c
3d
3e
3e
3 f
3g
3h
3i
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3CH2
CH3
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4a
4b
4a
5aa
5ba
5ba
5ca
5da
5ea
5ea
5 fa
5ga
5ha
5ia
69
72
72
72
72
72
72
72
72
85
85
72
85
120
92
80
97
81
83
47
78
87
80
70
75
65
67
82
19:1
19:1
15:1
13:1
21:1
22:1
15:1
8:1
12:1
5:1
3:1
4:1
19:1
16:1
84
87
83
83
81
88
83
72
80
80
77
92
85
85
4-CH3C6H4
4-CH3C6H4
3-CH3C6H4
2-CH3C6H4
4-CH3OC6H4
4-CH3OC6H4
4-ClC6H4
2-furyl
E-cinnamyl
Ph(CH2)2
Cy
C6H5
C6H5
3[d]
4
5
6
7[d]
8
9
10
11
12[d]
13
14[e]
tron-withdrawing
substituent
resulted in a slightly lower stereo-
selectivity (Table 2, entry 8). Heter-
oaromatic aldehyde 3g gave prod-
uct 5ga with good d.r. and ee values
(Table 2, entry 9). The present
system is also applicable to both
a,b-unsaturated and aliphatic alde-
hydes, which delivered products in
3j
3a
3a
5ja
5ab
5aa
[a] The reaction was run with 10 mol% of Pd/La/1 complex and 4-bromophenol at À408C unless
otherwise noted. Cy=cyclohexyl. [b] Yield of product isolated after column chromatography.
[c] Determined by 1HNMR analysis of the crude reaction mixture. [d] Reaction was run at À308C.
[e] Reaction was performed with 5 mol% of Pd/La/1 complex and 4-bromophenol. [f] Determined for
anti-5.
Angew. Chem. Int. Ed. 2008, 47, 3230 –3233
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3231