Angewandte
Chemie
Table 3: Variation of the enal.
when a catalyst loading of 1 mol% was employed, the
reaction time had to be increased to 6 h (Table 1, entry 7).
The reaction was slowed by decreasing the amount of DBU
(entry 5) and reducing the amount of acetyl acetone to
1.1 equiv also led to a lower yield (Table 1, entry 6). There-
fore, most of the following experiments were conducted with
2 mol% of the catalyst and 1.5 equiv of the nucleophile.
To study the scope of the redox-activated Michael
addition, we tested various 1,3-dicarbonyl compounds in the
reaction with cinnamaldehyde (Table 2).[15] Symmetrical 1,3-
R1
R2
R3
Product
t [h]
Yield [%][a]
4-MeOC6H4
4-Me2NC6H4
4-MeOCOC6H4
2-NO2C6H4
2-thienyl
Me
H
H
H
H
H
H
H
H
H
Me
H
H
H
H
H
H
H
H
Me
H
7b
7c
7d
7e
7 f
7g
7h
7i
1.5
4.0
1.5
1.5
1.5
2.5
2.5
2.5
48.0
36.0
89
85
82
86
75
81
92
91
Table 2: Variation of the nucleophile.
iPr
=
CH CHCH3
Ph[b]
7j
7k
34[c,d]
73
Me[b]
[a] Yield of isolated product. [b] With 10 mol% catalyst and 15 mol%
DBU. [c] Combined yield of both diastereoisomers. [d] Diastereoisomer
ratio 1.5:1.
R1
R2
Product
t [h]
Yield [%][a]
Et
Ph
Ph
OEt
OiPr
OMe
OMe
OEt
Et
5b
5c
5d
5e
5 f
5g
5h
5i
8.0
4.0
2.5
2.5
2.5
2.5
2.5
4.0
8.0
12.0
89
82
Ph
Me
Me
Me
Et
86[b,c]
80
give 7i (91%). The a-substituted a,b-unsaturated aldehyde 1j
afforded a lower yield even when a higher catalyst loading
was used (!7j, 34%). In this reaction both diastereoisomers
were formed (d.r. 1.5:1.0).[16] Importantly, the reaction of the
b-disubstituted enal 1k provided dihydropyranone 7k in 73%
yield, showing that our novel method can be used to build up
quaternary carbon centers. Note that a higher catalyst loading
and a longer reaction time were necessary in that case.
Importantly, we also found that a high stereoselectivity
can be achieved using the chiral b-ketoamide 3l as the
nucleophile in the reaction with para-methoxy cinnamalde-
hyde 1b (Scheme 3).[17]
74
81
76
81
76
51
=
CH2CH2CH CH2
Ph
4-MeOC6H4
Me
OMe
5j
5k
N-morpholinyl[d]
[a] Yield of isolated product. [b] Combined yield of both regioisomers.
[c] Regioisomer ratio 8:1. [d] With 5 mol% catalyst and 1.1 equiv DBU.
diketones bearing aliphatic and aromatic substituents
afforded the corresponding products 5b and 5c in good
yields. With these bulkier diketones the reaction was slower.
When the unsymmetrical 1,3-diketone 3d was used as a
nucleophile, both product regioisomers were formed. How-
ever, good regioselectivity was achieved (ratio 8:1). Probably
for steric reasons, the keto function next to the less bulky
methyl substituent preferably acted as the nucleophile in the
cyclization reaction to give 5d. We found that b-keto esters
are also efficient nucleophiles for NHC-catalyzed oxidative
1,4-additions (!5e–j, 74–81%). b-Alkyl b-keto esters
reacted significantly faster than b-aryl b-keto esters. The
reaction with b-keto amide 3k was slower and more catalyst
(5 mol%) and a stoichiometric amount of base were required
(!5k, 51%).
Scheme 3. Diastereoselective Michael addition using a chiral b-keto
amide.
We next examined the NHC-catalyzed Michael addition
of acetyl acetone 3a with various enals 1b–k (Table 3).
Electron-rich as well as electron-poor para-substituted cinna-
maldehyde derivatives provided the corresponding dihydro-
pyranones in high yields (!7b-d, 82–89%). ortho-Nitro
cinnamaldehyde underwent smooth redox-mediated addi-
tion/cyclization with acetyl acetone (!7e, 86%) and a
heteroaryl-substituted enal also worked well (!7 f, 75%).
The reaction with crotonal was efficient (!7g). With the
sterically more hindered isopropyl enal 1h, dihydropyranone
7h was isolated in excellent yield (92%). Regioselective
addition of 3a at the b carbon was achieved with dienal 1i to
The suggested catalytic cycle is depicted in Scheme 4. The
reaction between enal 1 and carbene 2 generates the electron-
rich enaminol A. Two-electron oxidation and deprotonation
of A with oxidant 4 provide the redox-activated Michael
acceptor C.[8] Conjugate addition of the likely deprotonated
1,3-dicarbonyl compound affords F. Proton transfer (!G)
and cyclization eventually lead to dihydropyranone 5 with
liberation of carbene 2.
To exclude other possible mechanisms we performed
control experiments. The reaction might also proceed by
kinetic O acylation of the enolate with C to form enol esters
Angew. Chem. Int. Ed. 2010, 49, 9266 –9269
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim