COMMUNICATION
well in toluene at À458C to afford the Mannich-type adduct
in good yield with moderate enantioselectivity (Table 1,
entry 1). It was found that the 3,5-disubstituted benzenes at
Substrate scope of b-dicarbonyl compounds was then ex-
amined. The malonates with primary alkyl chains, Er, Pr,
and Bu, worked well, and high enantioselectivities (95–
98% ee) were obtained (Table 2, entries 1–3). The allyl sub-
Table 1. Optimization of the reaction conditions.[a]
Table 2. Substrate scope of b-dicarbonyl compounds.[a]
Entry
3
x
Solvent
Additive
Yield [%][b]
ee [%]
Entry
2 (R1)
x
3
Solvent
Yield[b] [%]
ee [%]
1
2
3
4
5
6
3a
3b
3c
3d
3e
3 f
3g
3g
3g
3g
3g
3g
3g
3g
3g
3g
3g
20
20
20
20
20
20
20
20
20
20
20
10
10
5
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
DCM
THF
THF
THF
THF
THF
THF
THF
THF
–
–
–
–
–
–
–
–
76
89
83
78
78
89
62
trace
41
66
86
47
78
53
63
86
60
61
52
37
60
90
90
92
-
91
97
96
91
96
87
87
98
95
1
2
3
4
2b (Et)
2c (Pr)
2d (Bu)
2e (Allyl)
2 f (Bn)
2g (iPr)
2h (Ph)
2i (Cl)
5
5
5
5
5
10
10
10
10
20
3g
3g
3g
3g
3g
3h
3h
3e
3h
THF
THF
THF
THF
80
72
78
60
25
47
40
72
50[f]
96
95
98
95
87
89
10
22
79[g]
5
THF
6[c]
7[c]
8[c,d]
9[e]
toluene
toluene
toluene
toluene
7
8c
9
10
–
–
11[d]
12[d]
13[d,f]
14[d,g]
15[d,g,h]
16[d,g,h,i]
17[e,g,h,j]
MS 3ꢁ
MS 3ꢁ
MS 3ꢁ
MS 3ꢁ
MS 3ꢁ
MS 3ꢁ
MS 3ꢁ
[a] The reaction of 1 (0.600 mmol) with 2 (0.720 mol) was conducted by
using chiral Zr complex prepared from Zr(OtBu) (x mol%) and
a
ACHTUNGTRENNUNG
BINOL derivative 3 (1.1x mol%) at À208C for 18 h in 0.25m in the pres-
ence of MS 3ꢁ unless otherwise noted. 1 was slowly added during the re-
action time (18 h). [b] Yield of isolated product. [c] Without MS 3 ꢁ.
[d] Dimethyl ester was used. [e] Benzyl 2-oxocyclohexanecarboxylate (5)
was used as a substrate. The reaction was conducted at 08C for 24 h with-
out slow addition using a zirconium catalyst (20 mol%). [f] As a diaster-
eomixture. Diastereomer ratio was 3:1. [g] ee of major diastereomer.
5
5
2
[a] The reaction of 1 (0.600 mmol) with 2a (R1 =Me, 0.600 mmol) was
conducted by using chiral Zr complex prepared from Zr(OtBu)4
a
ACHTUNGTRENNUNG
(x mol%) and BINOL derivative 3 (1.1x mol%) at À458C for 24 h in
0.25m unless otherwise noted. The concentration was based on 1.
[b] Yield of isolated product. [c] Zr
ACHTUNGERTN(NUNG OiPr)4·iPrOH complex was used in-
stead of Zr(OtBu)4. [d] 18 h. [e] 46 h. [f] 0.4m. [g] 0.8m. [h] 1.2 equivalents
of 2a were added. [i] 1 was slowly added over 10 h. [j] 1 was slowly added
over 20 h.
ACHTUNGTRENNUNG
stituent also worked well (entry 4). On the other hand, the
benzyl substituent decreased the reactivity, although good
enantioselectivity was maintained (entry 5). When the malo-
nate bearing the iPr group was examined, good results were
obtained by using BINOL 3h (entry 6). The substrate con-
taining aromatic or halogen substituent did not work well
(entries 7, 8). The reaction of b-ketoester 5 was also investi-
gated, however selectivities were not sufficient (entry 9).
Next, deprotection of the obtained product was investigat-
ed. Oxidative cleavage of the PMP group using celium am-
monium nitrate (CAN) proceeded smoothly to afford the
desired primary amine in high yield under standard condi-
tions (in a water–acetonitrile mixture at room temperature)
(Scheme 2).[7,8]
the R2 parts did not increase the enantioselectivity signifi-
cantly (entries 2–4). On the other hand, it was revealed that
Br, CF3, and Me groups at the R2 parts were effective, and
high enantioselectivities were obtained (entries 5–7). Inter-
estingly, no reaction occurred when Zr
ACHTUNGTERN(NGNU OiPr)4 was used in-
stead of Zr(OtBu)4, suggesting that Brønsted basicity of the
ACHTUNGTRENNUNG
tert-butoxy group was very important in this catalyst system
(entry 8). When the catalyst prepared from 3g was used, ex-
cellent enantioselectivity (97% ee) was obtained when THF
was used as a solvent (entry 10). For improvement of the
yield, addition of MS 3 ꢁ was effective to afford the desired
Mannich-type product in 86% yield with 96% ee (entry 11).
We then investigated the reduction of the catalyst loading.
In the reaction using 10 mol% of the catalyst, significant de-
crease of the yield was observed; however, the reactivity
was improved by conducting the reaction in higher concen-
tration (entries 12, 13). While 5 mol% of the catalyst also
worked at higher concentration, both yield and selectivity
decreased (entry 14). Use of a slight excess amount of malo-
nate improved the yield (entry 15). Finally, it was found that
slow addition of the iminoester was very effective to im-
prove the enantioselectivity dramatically, and that the high-
est enantioselectivity (98% ee) was obtained (entry 16).
Under the reaction conditions, even 2 mol% of the catalyst
also worked well to give the product in moderate yield
while keeping a high enantioselectivity (entry 17).
Scheme 2. Deprotection of p-methoxyphenyl group.
A proposed catalytic cycle was shown in Scheme 3. First
the malonate interacts with the zirconium catalyst in a bi-
dentate fashion to form the active zirconium enolate (A) se-
lectively. The iminoester then coordinates to the zirconium
and is activated. Finally, the zirconium enolate attacks the
imine part of the iminoester to afford the desired product.
Formation of the zirconium enolate was confirmed by 1H
494
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 493 – 495