Organic Letters
Letter
a
Scheme 1. Modern Strategies for Accessing Alkynyl
Coumarins
Table 1. Evaluation of Reaction Parameters
b
entry
catalyst
solvent
yield (%)
1
2
3
4
5
6
7
8
Et3N
DMAP
DBU
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
MeCN
DCE
83
92
trace
37
32
25
16
trace
33
18
34
78
28
52
8
DABCO
Sparteine
Na2CO3
NaHCO3
NaOH
DMAP
DMAP
DMAP
DMAP
DMAP
DMAP
DMAP
DMAP
DMAP
c
9
10
11
12
13
14
15
16
17
DMF
EtOH
DMSO
toluene
MeCN
MeCN
MeCN
d
e
89
61
f
a
Reaction conditions: 4-methyl-2-(3-phenyl-1-(pyrrolidin-1-yl)prop-
2-yn-1-yl)phenol (1b) (0.2 mmol), 1,1′-(2-oxopropane-1,3-diyl)bis-
(pyridin-1-ium) bromide (2) (0.24 mmol), and catalyst (20 mol %)
b
in solvent (3 mL) at 80 °C for 5 h. Isolated yields were reported.
c
d
e
f
The catalyst loading was 10 mol %. At 50 °C. At 100 °C. For 2 h.
We initially explored the new reagent, dipyridinium ylide 2,
as a two-carbon feedstock for assembling the coumarin
scaffold. Thus, we employed propargylic amine 1b and 2 as
model substrates for this organocatalytic reaction (Table 1). In
the presence of 20 mol % organocatalysts, for instance Et3N
and DMAP, the reaction proceeded smoothly to give an up to
92% product yield (entries 1−2). Nevertheless, DBU,
DABCO, and sparteine were found to be less successful
(entries 3−5). Apart from the organic bases, commonly used
Na2CO3, NaHCO3, and NaOH did not facilitate this
transformation (entries 6−8). MeCN was the best solvent of
choice among other common organic solvents screened, e.g.,
DCE, DMF, EtOH, DMSO, and toluene (entries 10−14 vs 2).
There was no extra benefit of product yield at elevated reaction
temperature, e.g., 100 °C (entry 16 vs 2).
Having the optimized reaction conditions in hand, we next
explored the substrate scope (Scheme 2). In general, the
coumarin products were obtained in good-to-excellent yield.
Particularly noteworthy is that the −Br and −Cl groups
remained intact during the course of the reaction. This
outcome allows these products to be further modified by
established cross-coupling strategies at a later stage. There was
no significant electronic effect of the substituents displayed at
the phenolic ring (products 3a−3d and 3e−3h). Likewise, the
electronic property of the alkynyl arenes was also insignificant
(products 3i−3l vs 3m−3p). It is important to show that the
sterically hindered phenolic fragment did not affect the
efficiency of the cyclization (products 3q−3s). Even the
highly sterically bulky tert-butyl group was well-tolerated
(product 3r). There were no steric influences at the alkynyl
arenes as well (products 3t and 3w). Dihalo-substituents at
either phenolic or alkynyl arene rings were also compatible
(products 3s−3u). The product 3t was unambiguously
characterized by single crystal X-ray crystallography. The
highly electron-withdrawing nitro group did not affect the yield
of coumarin scaffold assembly (products 3v and 3x). The
possible gram-scale synthesis showed the potential practic-
ability for large-scale preparation of substituted coumarins.
In addition to aromatic alkyne substituents, the applicability
of other alkenyl-, thienyl-, and alkyl-containing substrates were
examined (Scheme 3). It is worth noting that the conjugated
enyne moiety was also well-suited in this catalyst system
(product 3y). There was no deleterious effect of a heterocyclic
substrate under these reaction conditions (product 3z).
Not only the alkynyl coumarin scaffold has rich application
in material sciences, the corresponding alkenyl coumarin
skeleton also displays unique photophysical properties.31
Common modular assembly of these alkenyl coumarin units
relies on the palladium-catalyzed Suzuki−Miyaura and Heck
couplings of coumarin sulfonates/bromides with either
alkenylboronic acids32 or alkenes,33 respectively. The inherent
limitation of these existing protocols would be the incompat-
ibility of −Br or −Cl groups. Indeed, it would be highly
attractive if we can develop an organocatalytic method
particularly fit for moderate functional group tolerance. Our
further attempts of using dipyridinium ylide 2 as the acyl
carbene surrogate led us to have a variety of halo-containing
alkenyl coumarins (Scheme 4). There was no significant
substrate electronic effect with regard to the desired product
yields (products 5b−5f). The steric effect, where the −Br
group at the ortho-position to the phenolic group, was
insignificant (product 5g). It is noteworthy to show that this
catalyst system displayed entire compatibility with −Br and
−Cl groups. Thus it exhibits rich potential for subsequent
functionalization using cross-coupling technology.
B
Org. Lett. XXXX, XXX, XXX−XXX