Angewandte
Chemie
yield of isolated d-product 4a could be raised up to 92% while
still maintaining a high enantioselecitivity (89% ee; Table 1,
entry 12).
An extensive screen of other chiral amine catalysts
showed that cinchonine-derived catalyst D gave a slightly
higher ee value of 90%, albeit with a lower yield of 4a
(Table 1, entry 7). Catalyst E, which has no double bond,
afforded 4a in low yield and with a low ee value (Table 1,
entry 8 versus 6). The opposite enantiomer of the d product
was provided with a very low ee value by employing
quinidine-derived primary amine catalyst F under the same
reaction conditions (Table 1, entry 9). It was found that the
TFA salt of (1R, 2R)-1,2-diphenylethane-1,2-diamine B was
not able to promote the reaction. Proline (A) did not catalyze
the reaction of 1a with 2a either. The acid additives had an
obvious effect on the reactivity of the catalyst and enantio-
selectivity of the nucleophilic substitution. TFA exhibited
specific reactivity for the reaction. If TSA was used as an acid
additive, the reaction provided only 18% of 4a with 63% ee
(Table 1, entry 11). In the absence of any Brønsted acid as a
co-catalyst, no product was observed and starting materials
were recovered.
Optimal results for d selectivity with high enantioselec-
tivity were eventually obtained in the solvent mixture
containing equal amounts of iPrOAc and THF in the presence
of 20 mol% of 9-amino-9-deoxyepiquinine (C) in combina-
tion with TFA (40 mol%), which allowed the reaction to
proceed smoothly within 5 days at 308C. Under the optimal
reaction conditions we investigated the scope of d products
with different combinations of indoles and a variety of cyclic
MBH alcohols. The results are summarized in Table 2. In each
case, the corresponding d products were obtained in good to
excellent yields as single regioisomers. In general, for indole
2a, variation in the MBH alcohols was well tolerated. For
para-substituted MBH alcohols (1b–1g; Table 2, entries 2–7),
regardless of having electron-donating or electron-withdraw-
ing groups, gave good to excellent ee values (83–93%) for
d products. For meta-substituted MBH alcohols (1h–1k;
Table 2, entries 8–11), high enantioselectivities (88–92% ee)
and moderate to good yields (71–86%) were obtained.
However, no reaction was observed with cyclohexenone-
derived MBH alcohols and indole 2a. The MBH alcohol with
an alkyl substituent instead of an aryl one did not afford the
product either, which indicated that the benzyl alcohol
structure was necessary for the reaction of MBH alcohols
with indoles. For indole derivatives, 5-bromoindole (2e), 5-
methoxylindole (2 f), 5-methylindole (2g), and 6-methylin-
dole (2h) gave high yields with good enantioselectivities
(Table 2, entries 15–18). Exceptionally, the reaction of 2-
methylindole (2b) with 1a provided a mixture of d and
g product (d/g = 65:35) with a high enantioselectivity of
d product 4b (93% ee; Table 2, entry 12). Whereas 2-phenyl-
indole (2d) still exhibited exclusive d regioselectivity with a
lower enantioselectivity (Table 2, entry 14). The poor regio-
selectivity of 2-methylindole (2b) could be ascribed to its high
nucleophilic character compared to simple indole.[13] N-
Methylindole (2c) was less reactive under the reaction
conditions. By increasing the reaction temperature to 408C,
2c afforded the d-product 4m in 70% yield with only 47% ee.
Scheme 3. Structures of the chiral amine catalysts.
Table 1: Selected screening studies of organocatalytic nucleophilic
substitution of cyclic MBH alcohol 1a with indole 2a.[a]
Entry
Catalyst[b]
Solvent
Yield of
ee of
4a [%][c]
4a [%][d]
1
2
3
4
5
6
7
8
C
C
C
C
C
C
D
E
CH2Cl2
Et2O
THF
EtOAc
50
59
28
70
77
45
35
35
35
70
18
92
66
52
85
49
64
89
90
80
-47
75
63
89
iPrOAc
iPrOAc/THF (1:1)
iPrOAc/THF (1:1)
iPrOAc/THF (1:1)
iPrOAc/THF (1:1)
iPrOAc/THF (1:1)
iPrOAc/THF (1:1)
iPrOAc/THF (1:1)
9
F
10[e]
11
12[g]
C
C[f]
C
[a] Reaction conditions: unless otherwise noted reactions were per-
formed with 0.2 mmol of 1a, 0.2 mmol of 2a, 0.04 mmol of catalyst C,
and 0.08 mmol of TFA in 2 mL of solvent at 308C for 3 days. [b] With TFA
as an acid additive. [c] Yield of isolated product. [d] Determined by HPLC
on a chiral stationary phase. [e] At 508C. [f] TSA was used as an acid
additive. [g] 0.4 mmol of 1a; the reaction was carried out over 5 days and
the yield is based on 2a. THF=tetrahydrofuran, TSA=p-toluenesulfonic
acid.
Scheme 1), while d selectivity is unusual and unexpected. The
reaction in Et2O provided 4a in 59% yield and 52% ee, but no
g product was detected (Table 1, entry 2). The examination of
solvent effects revealed that in THF the reaction afforded the
d-product 4a with good enantioselectivity (85% ee) but low
yield and no g product was observed (Table 1, entry 3);
whereas ethyl acetate significantly facilitated the reaction
and produced the d-product 4a in higher yield but with a
relatively lower ee value (Table 1, entry 4). In isopropyl
acetate, the reaction provided 4a in 77% yield and 64% ee
(Table 1, entry 5). Other solvents such as toluene, acetonitrile,
methanol, and N,N-dimethylformamide were also tested;
they were less effective for the reaction of 1a with 2a (see the
Supporting Information). Based on these experimental
results, a solvent mixture of iPrOAc/THF (1:1) was employed
as the reaction medium. The enantioselectivity of the d-
product 4a was increased up to 89% ee, but the yield still
remained low (compare Table 1, entrie 6 versus 3 and 5).
Although the yield could be increased to 70% by raising the
reaction temperature to 508C, unfortunately the ee value
decreased to 75% (Table 1, entry 10). To reach both high
yield and enantioselectivity, we changed the ratio of 1a/2a
from 1:1 to 2:1, and prolonged the reaction time to 5 days, the
Angew. Chem. Int. Ed. 2010, 49, 7294 –7298
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7295