2
Tetrahedron
reported.9 In 2015, Xuan and coworkers described oxidative
Table 1. Optimization of direct addition of 4-quinazolinone
(1a) with 4-vinylpyridine (2a).a
coupling of 2-aryl-4-quinazolinones with olefins through C-H
activation/aza-Michael reaction strategy.9a Subsequently, Peng
found rhodium could also promote these tandem reactions
successfully.9b Interestingly, Engle reported palladium-catalyzed
directed alkene hydroamination between 2-pyridone derivatives
and olefins in 2017.9c Recently, Hou reported the first aza-
Michael addition of 2-hydroxypyridines with α, β-unsaturated
1,4-dicarbonyl compounds with high N-selectivity in 2019.10
N
O
N
N
OH
O
N
catalyst
N
N
N
+
+
solvent, T
N
N
3b
3a
1a
2a
not observed
entry
1
catalyst
solvent
DMSO
T(℃)
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
110
80
yield%b
54
43
58
17
55
42
34
56
91
79
19
43
71
23
67
72
56
0
On the other hand, vinylazaarenes have been successfully
employed as novel Michael acceptor by Lam,11 Wang,12
Harutyunyan,13 Terada14 and Jiang15 in recent years. A series of
nucleophiles such as aldehydes, azoles,16 amino acid esters,17 and
alkoxyamine18 have been utilized to react with the novel Michael
acceptor.19 We envisioned whether this methodology could be
smoothly transferred to derivatization of tautomerizable
heterocycles. Herein, we report the first Sc(OTf)3 catalyzed aza-
Michael addition reaction between quinazolinones and
vinylazaarenes.
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
Cu(OAc)2
FeCl3
2
1,4-dioxane
DMF
3
4
THF
5
EtOAc
DCE
6
7
CH3CN
neat
8
To begin, 4-hydroxyquinazoline (4-quinazolinone) 1a was
selected as a model substrate with two distinct tautomerizable
nucleophilic centers (–OH and –NH) to react with 4-
vinylpyridine 2a (Table 1). To our delight, N-alkylated
quinazolinone product 3a was obtained in 54% yield with 10
9
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
10
11
12
13
14
15
o
mol% Sc(OTf)3 catalyst in 3 mL DMSO at 110 C under an N2
Pd(OAc)2
Pd(PPh)4
ZnBr2
atmosphere without observing any O-alkylated product (Table1,
entry 1). The studies showed that the solvent had a great impact
on this reaction. It proved that only 17% yield was observed
when the reaction was carried out in THF (entry 4). After
screening of solvent, the product 3a was obtained in 91% yield
with toluene as solvent (Table1, entry 9). Next, a series of Lewis
acids were tested for the reaction. Cu(OAc)2, FeCl3, Pd(OAc)2,
Pd(PPh)4 or ZnBr2 could also promote the reaction with relatively
lower yield (entry 10-14). It should be noted that Brφnsted acid
could also promote the transformation (entry 15). Considering
the corrosiveness of HOTf, it was not chosen for catalyzing the
reaction. In the absence of catalyst, no reaction was occurred
(entry 19). A systematic survey of catalysts revealed that
Sc(OTf)3 could promote the reaction most effectively. The yield
decreased apparently when the catalyst loading was lowered to 5
mol%. (entry 16) Decreasing reaction temperature results in
much lower yield (entry 17). Moreover, the reaction could not
proceed at all when the temperature was lowered to 60 ℃. (entry
18) Therefore, the optimal reaction condition was confirmed as
follow: a solution of 1 (1 equiv), 2 (2 equiv), 10 mol% Sc(OTf)3
in toluene was stirred for 12 h at 110 oC with a N2 balloon (entry
9).
HOTf
16c
17
18
19
a
Sc(OTf)3
Sc(OTf)3
Sc(OTf)3
-
60
110
0
Reaction conditions: 1a (0.5 mmol), 2a (1 mmol), 10 mol%
catalyst, in 3 mL solvent at 110 oC for 12 h.
b isolated yield.
c 5 mol% of Sc(OTf)3 was used as the catalyst.
or meta-position could react with 4-hydroxyquinazoline
successfully in 51%-87% yield. It appeared those substrates with
electron-donating groups resulted in lower yields (3l, 3m), which
may be attributed to the lower reactivity of these compounds. We
then explored whether this reaction could be applied for
heterocycles other than pyridine. Gratifyingly, heterocycles such
as pyrimidine and benzoxazole were well tolerated (3n, 3o) with
good yields. But the reaction between 1a and 2-vinylquinoline
afforded product 3p in 26% yield. Next, it was found that the
reaction between 1a and (E)-2-(prop-1-en-1-yl)pyridine couldn’t
occur, probably due to its lower reactivity (3q).
With the optimal condition in hand, the scope of the reaction
between 4-hydroxyquinazoline 1a and various vinylazaarenes
was probed. (Table 2) A wide range of readily accessible
vinylazaarenes underwent this reaction with 1a in excellent
chemo selectivity. Generally, 2-vinylpyridines with electron-
deficient or electron-rich groups afforded products 3a-3f in
moderate to excellent yields (73-89%). In the case of para-
substituted substrate 3f, lower yield was observed compared to
3d and 3e. Additionally, the reaction of 4-hydroxyquinazoline
with 4-vinylpyridine resulted in 91% yield (3g). Next, we
examined the reactions for 4-vinylpyridines with different
functional groups. It turned out that 4-vinylpyridines with
halogen atoms (3h-3k) such as -Cl, -Br, -F on the ortho-position
Subsequently, the scope of 4-hydroxyquinazolines was
examined for the reaction with 4-vinylpyridine as a standard
substrate and the results are shown in Table 3. 4-
hydroxyquinazolines with different substituents could react well
to produce 5 in moderate to high yields with excellent chemo
selectivity. It was found that these 4-hydroxyquinazolines with
electron-deficient groups reacted well to afford the adduct
product in 62-95% yields (Table 3, 5a-5i). Besides, the position
of substituent groups on the phenyl didn’t affect the reaction
apparently (5b-5d). 4-hydroxyquinazolines with electron-rich
groups on the phenyl afforded 55-74% yield (5j, 5k), indicating