.
Angewandte
Communications
organocatalysts.[18] Jacobsen and co-workers recently discov-
ered that the Povarov reaction could be promoted effectively
by the combined use of bifunctional urea/thiourea catalysts
with a strong Brønsted acid.[19] We thus examined the model
reaction in the presence of acid additives (Table 3, entries 12–
14). AcOH (20 mol%) appeared to be the most effective
additive and gave the product with an excellent 89% ee in
a moderate yield of 61% (Table 3, entry 13). By screening
mixed solvents, we found that a mixture of iPrOH with DMF
(4:1) afforded 3aa with excellent yield (98%) and enantio-
selectivity (92%, Table 3, entry 18). Decreasing the reaction
temperature resulted in a slightly decreased yield and
prolonged reaction time (Table 3, entry 21). Furthermore,
the catalyst loading could be decreased to 5 mol% to afford
3aa in 86% yield with consistent enantioselectivity, but
prolonged reaction time (Table 3, entry 22). Further addition
of 20 or 40 mol% of AcOH to the system shortened the
reaction time (Table 3, entries 23 and 24), but not in the case
where the catalyst loading was lowered to 5 mol% (Table 3,
entry 25).
After we established the optimal reaction conditions, the
generality of this reaction was probed by using a variety of
gem-diols 1a–1l as aldol donors (Table 4). In the presence of
Cat 6 (10 mol%) and AcOH (20 mol%), the aldol reaction of
N-benzyl isatin 2a with aryl-substituted 2,4,4,4-tetrafluoro-
3,3-dihydroxy-1-butan-1-ones 1a–1h proceeded smoothly in
iPrOH/DMF (4:1) at room temperature, giving the desired
products 3aa–3ha in admirable yields (91–99%) with excel-
lent diastereoselectivities (99:1) and stereoselectivities (83–
98% ee), regardless of the positions and electronic nature of
the substituents on the phenyl rings (Table 4, entries 1–8).
Similar results were also obtained when R1 was replaced by
the heterocyclic 2-furyl and 2-thienyl groups (Table 4,
entries 9,10). The reaction could also be carried out with
aliphatic gem-diols, though more sluggishly, affording the
corresponding products in moderate yields, but still with high
diastereo- and stereoselectivities (Table 4, entries 11–12). The
absolute configuration of the major diastereomer of 3aa was
determined as 3S,1’S by X-ray analysis (Figure 1).[20]
Figure 1. X-ray crystal structure of the major diastereomer of 3aa.
Thermal ellipsoids are set at 30% probability.
Next, we probed the reaction scope with a range of N-
benzyl isatins 2b–2i as aldol acceptors (Table 5). With 1a as
the substrate, various isatins with different substituents on the
aromatic ring, including halogens and electron-donating
groups, are suitable for this reaction. Excellent yields (91–
Table 5: Generality of the reaction with aldol acceptors 2b-2i.[a]
Entry R1/R2
Time [h] Yield [%][b] d.r. [%][c] ee [%][d]
1
2
3
4
5
6
7
8
9
H (1a)/5-F (2b)
16
16
18
16
16
99 (3ab)
99 (3ac)
99 (3ad)
99 (3ae)
99 (3af)
98 (3ag)
91 (3ah)
95 (3ai)
98 (3cb)
99 (3eb)
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
92
91
92
86
92
98
91
83
83
93
H (1a)/6-Br (2c)
H (1a)/5-Br (2d)
H (1a)/7-Cl (2e)
H (1a)/5-Me (2 f)
Table 4: Generality of the aldol reaction.[a]
H (1a)/5-OMe (2g) 16
H (1a)/4-Br (2h)
H (1a)/5,7-Me2(2i)
4-F (1c)/5-F (2b)
24
24
16
10
4-OMe (1e)/5-F (2b) 19
Entry R1
Time [h] Yield [%][b] d.r. [%][c] ee [%][d]
[a] Reactions were performed with various substituted b-nitroolefins 2b–
2i (0.2 mmol) with 2,4,4,4-tetrafluoro-3,3-dihydroxy-1-arylbutan-1-one
1 (0.3 mmol) and Cat 6 (0.02 mmol) in iPrOH/DMF (4:1, 2.5 mL).
[b] Yields of isolated products. [c] Determined by 19F NMR spectroscopic
analysis of unpurified reaction products. [d] Determined by HPLC using
a chiral stationary phase.
1
Ph (1a)
16
16
16
16
99 (3aa)
97 (3ba)
99 (3ca)
95 (3da)
96 (3ea)
91 (3 fa)
99 (3ga)
97 (3ha)
99 (3ia)
99 (3ja)
43 (3ka)
67 (3la)
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
99:1
92
93
91
83
95
93
88
91
92
98
83
83
2
3
3-MeC6H4 (1b)
4-FC6H4 (1c)
4
3,4-F2C6H4 (1d)
5
4-MeOC6H4 (1e) 36
6
7
8
9
2-MeC6H4 (1 f)
4-BrC6H4 (1g)
4-ClC6H4 (1h)
2-furyl (1i)
36
20
20
28
28
60
60
99 %), high diastereoselectivities (99:1) and stereoselectivi-
ties (83–98% ee) were obtained (Table 5, entries 1–8).
Furthermore, we investigated the reaction of (4-fluoro-
phenyl)-(1c) and (4-methoxyphenyl)-3,3-dihydroxybutan-1-
one (1e) with isatin 2b and found that the electronic
properties of the substituents had a negligible impact on the
reactivity (Table 5, entries 9–10).
To study the plausible mechanism of the reaction, we
further synthesized two substrates to verify that the asym-
metric aldol addition is not a stepwise process in which release
10
11
12
2-thienyl (1j)
Me (1k)
cyclopropyl (1l)
[a] Reactions were performed with 2a (0.2 mmol) with various substi-
tuted 2,4,4,4-tetrafluoro-3,3-dihydroxy-1-butan-1-ones 1a–1l (0.3 mmol)
and Cat 6 (0.02 mmol) in iPrOH/DMF (4:1, 2.5 mL). [b] Yields of
isolated products. [c] Determined by 19F NMR spectroscopic analysis of
unpurified reaction products. [d] Determined by HPLC using a chiral
stationary phase.
5568
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 5566 –5570