Communication
and amine donor iPr NEt failed to give the desired products
[
a]
2
Table 1. Screening of the reaction conditions.
whether in the presence of Ru(bpy) Cl or FIrPic. These primary
3
2
observations are quite different with [2+2] cycloadditions of
[
6]
aryl enones promoted by photo-induced electron transfer.
Therefore, an alternative energy-transfer mechanism should be
involved for the reaction.
[
b]
To test the viability of this method, we next turned our at-
tention to [2 ++ 2] cycloaddition of coumarin-3-carboxylate 1a
with additional acrylamide derivatives. A variety of N,N-disub-
stituted acrylamides are good reaction partners for the crossed
[2+2] cycloaddition (Table 2, entries 1–7). It was found that
yields of [2+2] cycloadducts were sensitive to the steric hin-
drance of alkene moieties of the acrylamides. For instance, the
methyl group at the 2-position of acrylamide significantly de-
creased the production of cycloadditives (Table 2, entry 2). The
broad utility of this transformation is presented by various
amino moieties of the acrylamides, such as methylaniline, di-
ethylamine, piperidine, morpholine, piperazine, and piperidin-
2-one. Moreover, acrylate esters and aliphatic enone also par-
ticipate and provide the desired [2+2] cycloadducts (Table 2,
entries 8–10).
Entry
Solvent
Conditions
Yield [%]
[
c]
1
2
3
4
5
6
7
8
9
CH
3
CH
3
CH
3
CH
3
CH
3
CH
3
CN
CN
CN
CN
CN
CN
Ru(bpy)Cl
2
N.R.
N.R.
Ir(ppy)
FIrPic
3
83 (1.1:1)
51 (1:1)
N.R.
[Ir(dtbbpy)(ppy)
FIrPic
–
2
]PF
no light
6
N.R.
toluene
CH Cl
FIrPic
FIrPic
FIrPic
FIrPic
FIrPic
FIrPic
FIrPic
FIrPic
FIrPic
76 (1.6:1)
74 (1.6:1)
76 (1.7:1)
38 (1.8:1)
80 (1:1)
2
2
acetone
THF
DMF
1
0
1
1
1
1
1
1
1
2
3
4
5
6
DMSO
70 (1:1)
CH
3
CH
3
CH
3
CH
3
OH
CN
CN
CN
56 (2.7:1)
71 (1.4:1)
LiBF
4
[
d]
d]
[e]
LiBF4, iPr
LiBF4, iPr
2
2
NEt
NEt
N.D.
[
[e]
Ru(bpy)Cl
2
N.D.
Experiments exploring the generality of the crossed reaction
with respect to the coumarin analogs are summarized in
Table 3. A range of coumarin-3-carboxylates with different sub-
stituents on the aromatic ring are well-tolerated under the re-
action conditions, which generate various cyclobutabenzopyr-
ancarboxylate esters in isolated yields ranging from 60 to 91%
[
a] Reaction conditions: a mixture of 1a (0.20 mmol), 2a (0.24 mmol), and
(
Table 3, entries 1–7). The process appears to be general with
photocatalyst (0.5 mol%) in solvent (2 mL) was irradiated with 3 W blue
LEDs at room temperature for 24 h. [b] Yield of isolated product. The
ratio in brackets is the ratio of 3a:4a. [c] N.R.=no reaction. [d] 2 equiv of
the steric effect of the ester group, as demonstrated by the ef-
ficient conversion of the tert-butyl ester to the corresponding
cycloaddition product (Table 3, entry 8). In addition, 3-acetyl-
substituted coumarin 1j and quinolone-3-carboxylates 1k–1o
4 2
LiBF and 4 equiv of iPr NEt. [e] N.D.=not detected.
[
13]
are also amenable to the [2+2] cross-cycloaddition, indicat-
ing that the substrate scope of the visible-light-driven reaction
is remarkably broad. It was interesting that the reaction of qui-
nolone-3-carboxylates displayed a significant preference for
the exo-products (Table 3, entries 10–14), which can be ex-
plained by the steric influence on the intermediate 1,4-diradi-
cal species formed in the course of the [2+2] photocycloaddi-
atmospheres under 3 W blue LEDs (l=450 nm) irradiation. As
shown in Table 1, Ru(bpy) Cl and Ir(ppy) , which have been ex-
3
2
3
tensively explored as single-electron transfer photocatalysts,
fail to promote this reaction. To our delight, the desired 1-exo-
and 1-endo-substituted benzocyclobutapyranones (3a and 4a)
[12]
were isolated in 83% yield when FIrPic was employed as the
[
13a]
photocatalyst (Table 1, entry3). The stereochemistry of 3a and
tion.
1
4
a were determined by H NMR and X-ray crystallographic
The synthetic utility of the [2+2] cross-cycloaddition prod-
ucts was exemplified by ring-opening reactions (Scheme 1).
When a diastereomeric mixture of 1-exo- and 1-endo-substitut-
ed benzocyclobutapyranones (3a and 4a) was treated with
2.0 equivalents of dimethylsulfoxonium methylide in DMF at
room temperature, the desired tetrahydrodibenzofuran (7) was
obtained in 64% yield. Dibenzofurans have been found as
analysis (see Figure S3 in the Supporting Information). Another
Ir photocatalyst, [Ir(dtbbpy)(ppy) ]PF , can also catalyze the visi-
2
6
ble-light-driven [2+2] cycloaddition regardless of the de-
creased yield of [2+2] cycloaddition products (Table 1, entry 4).
Control reactions confirmed that no reaction occurs either in
the absence of light or the absence of photocatalyst (Table 1,
entries 5 and 6). Further solvent screening was performed and
the results showed that the cycloaddition process is not sensi-
tive to the reaction medium (Table 1, entries 7–13). For exam-
ple, although CH CN gave a higher yield, toluene, CH Cl , ace-
3
2
2
tone, DMF, and DMSO also are suitable solvents for this reac-
+
tion. In addition, LiBF , a Lewis acidic Li additive, which had
4
been used to activate the enone substrate towards one-elec-
tron reduction and stabilize the resulting radical anion species
[
3]
in Yoon’s system, decreased the yield of the cycloaddition
Table 1, entry 14). Moreover, combination of Lewis acid LiBF4
Scheme 1. Transformation of benzocyclobutapyranones 3a and 4a to tetra-
hydrodibenzofuran 7.
(
Chem. Eur. J. 2015, 21, 10326 – 10329
10327
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Liu, Qiang
