Table 1. Screening of the acids.
successful example of a pyrrolidine-based catalyst,
capable of effecting the highly enantioselective
asymmetric alkylation of aldehydes and overcomes
the fallacy of the failure of pyrrolidine-based cata-
[5,8a]
lysts in this elusive reaction.
With the optimized conditions in hand, we pro-
ceeded to examine the substrate scope using I/2,3,4-
trihydroxybenzoic acid (2–5 mol%) as the catalytic
system in the reaction of xanthydrol with different
aldehydes (Table 2, entries 1–9). Gratifyingly, in all
cases, the desired products were obtained in good
to high yields with excellent enantioselectivities
[
a]
[b]
[c]
Entry
Acids
–
x [mol%]
t [h]
Yield [%]
ee [%]
1
2
3
4
5
6
7
8
9
0
12
12
12
12
12
12
12
12
12
6
48
6
6
24
12
8
0
0
0
0
0
0
0
–
–
–
–
–
–
–
33
86
–
(
91–98% ee), thus making this methodology highly
InCl
FeCl
LaCl
In(OTf)
Zn(OTf)
La(OTf)
CAN
3
50
50
50
50
50
50
50
50
10
10
10
10
10
10
10
10
10
5
valuable in view of the biochemical and pharma-
3
[13]
3
ceutical importance of xanthene derivatives. With
propionaldehyde, bis(4-(dimethylamino)phenyl)me-
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
2
[14]
thanol 2b
gave the desired product 3j in 87%
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
3
yield and 91% ee (Table 2, entry 10). Given the
24
59
trace
trace
56
48
72
67
80
64
92
91
83
CuCl
CSA
2
vital importance of ferrocene-based chiral ligands
10
11
12
13
14
15
16
17
18
19
20
[15]
and catalysts in asymmetric catalysis and the key
C
6
H
5
CO
2
H
–
role of the indole skeleton in natural products and
TFA
p-TSA
79
94
93
95
90
94
95
95
96
[16]
pharmaceutical chemisty, we proceeded to inves-
tigate the reactions of ferrocene- and indole-based
substrates. The results are summarized in Table 3.
To our surprise, the catalytic system I/2,3,4-trihy-
droxybenzoic acid did not work for ferrocene- and
indole-based substrates. Fortunately, it was found
that I/TfOH (TfOH=trifluouromethanesulfonic
acid), on the other hand, was an effective catalytic
2-NO
3-NO
4-NO
2
2
2
-C
6
H
H
H
4
4
4
CO
CO
CO
2
2
2
H
H
H
-C
-C
6
6
4-CF
3
-C
6
H
4
CO
2
H
6
6
12
15
[
[
[
d]
e]
f]
(HO)
(HO)
(HO)
3
3
3
C
C
C
6
6
6
H
H
H
2
2
2
CO
CO
CO
2
H
2
H
2
H
2
[
a] Reactions were conducted with butyraldehyde (0.4 mmol), xanthydrol (0.1 mmol),
catalyst I (0.01 mmol), and acid in CH Cl
isolated product. [c] ee was determined by chiral HPLC on a chiral stationary phase.
d] 2,3,4-Trihydroxybenzoic acid. [e] catalyst I (5 mol%) was used. [f] Catalyst
2 mol%) was used. CAN=ceric ammonium nitrate.
2
2
(0.5 mL) at room temperature. [b] Yield of system for these two important substrates. For fer-
rocene-based substrates, high yields and ee were ob-
served (3k and 3l). In sharp comparison, for
Cozziꢀs system, 90 mol% of MacMillan’s catalyst
was necessary to drive the reaction to completion
and the product was obtained in low yield and mod-
[
(
I
[5b]
product (Table 1, entry 1). Subsequently, a wide variety of
commonly used Lewis acids were added in addition to I. It
was found that the commonly used Lewis acids such as
indium salts, lanthanide complexes, and zinc complexes are
all ineffective for this transformation (Table 1, entries 2–8).
erate ee. To our delight, for indole-based substrates, the I/
TfOH system can also catalyze reaction of substrate 2d
(without 2-substitution) with aldehydes to provide 3m and
3n in good yields (70% yield and 69% yield, respectively),
excellent enantioselectivities (97–99% ee), and excellent
diastereoselectivities (anti/syn=93:7 for 3m, anti/syn=92:8
for 3n). In strong contrast with Melchiorreꢀs report on the
asymmetric a-alkylation of aldehydes through vinylogous
iminium ion intermediates generated by indole deriva-
It is interesting to note that CuCl catalyzed the reaction to
2
afford the desired product in good enantioselectivity, albeit
in moderate yield (Table 1, entry 9). Next, we turned our at-
tention to examine if Brønsted acids can work well with or-
ganocatalyst I. During our screening of a different combina-
tion of Brønsted acids with I, it was found that good yields
and high enantioselectivities (90%–95% ee) could be ob-
tained for most of the electron-deficient benzoic acids
[17]
tives, substrate 2d gave the product in only 11% ee and
(
Table 1, entries 14–17). However, no reaction was observed
with camphorsulfonic acid (CSA) and benzoic acid (Table 1,
entries 10–11). Trifluoroacetic acid (TFA) and p-TSA can
also afford satisfactory results (Table 1, entries 12–13).
Among them, 2,3,4-trihydroxybenzoic acid was found to
give the best results and the catalyst loading of organocata-
lyst I and acid could be decreased to as low as 2 mol% with-
out compromising the reactivity and enantioselectivity
(
Table 1, entries 18–20). Therefore, our method provides the
Scheme 2. Transition-state model for asymmetric alkylation of aldehydes.
Chem. Asian J. 2011, 6, 2890 – 2894
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2891