2
J. Chen et al. / Tetrahedron Letters xxx (2018) xxx–xxx
prevous work
Ar
OEt(Bu)
O
Pd(OAc)
2
R
R
R
ArI(Br)
(a)
(b)
n-
n-
Bu NOAc/ Bu NBr
4
4
O
O
O
OH
Ar
OEt
O
Pd(OAc)
2
ArN BF4
2
R
CaCO , MeOH
3
O
OH
o
60 C, 2-5h
Ar
OEt
O
Pd(OAc) or PdO-Fe O
2
3
4
(c)
R
Ar
I
Ar
OTf(BF )
R
4
DMF or EtOH
O
O
OH
o
80 C, 5h
Present work
Ar
O
OEt
O
Pd(OAc)
2
R
R
ArI
(d)
H O
2
O
OH
Scheme 1. Palladium catalyzed Heck-arylation/cyclization of ortho-hydroxy cinnamate with various electrophilic coupling partners.
of an atom-economic and environmentally friendly method for the
construction of 4-arylcoumarins remains a challenging task, and
therefore merits further consideration. Herein, we would like to
disclose our realization of the synthesis of 4-arylcoumarins by
palladium-catalyzed Heck-arylation/cyclization of ortho-hydroxy
cinnamates with aryl iodides in water.
(73%, Table 1, entry 25). Control experiments have, however,
proven that, in the absence of catalyst or base, no reaction took
place (Table 1, entries 26,27). Overall, the catalytic protocol using
PdCl2(CH3CN)2 as catalyst with NaOAc as base in H2O at 100 °C
was found to be the most effective combination for obtaining an
excellent yield (Table 1, entry 23).
Under the optimized reaction conditions (Table 1, entry 23), we
decided to test a wide range of aryl iodides 2 to evaluate the scope
of the arylation/cyclization protocol. As shown in Table 2, aryl
iodides possessing electron-neutral, electrondonating or electron-
withdrawing groups were well tolerated and smoothly underwent
arylation/cyclization reactions to result in variously functionalized
4-arylcoumarins in good to excellent yields. Furthermore, the sub-
stitution patterns had no influence on the present transformation
reactions. For example, aryl iodides having a methyl group at the
para, meta, or ortho position, respectively, afforded the desired 4-
aryl coumarins 3b–d in excellent yields. Aryl iodide bearing two
methyl groups at meta positions was tolerated and gave the corre-
sponding coupling products 3e in 92%yields. Similarly, methoxy
group at the para, meta, or ortho position also provided the corre-
sponding arylation/cyclization products 3f–h in excellent yields.
Specifically, halogen substituents such as fluoro, chloro and bromo
were well tolerated, and gave the corresponding products 3i–n in
high to excellent yields. These functional groups are useful syn-
thetic handles for further derivatizations. Additionally, aryl iodides
bearing an electron-withdrawing group were well-tolerated under
the optimized reaction conditions. For example, aryl iodides bear-
ing strong EWG such as trifluoromethyl, acetyl, and carboxylic acid
methyl ester afforded the desired products 3o–q in high yields. It is
interesting to mention that aryl iodide bearing a free amino group
at para position was well tolerated, and gave the corresponding
products 3r in 67% yield. The 1-iodo-naphthalene was a good sub-
strate and gave the desired product 3s in 77% yield. The reaction
also proceeded well with heterocyclic coupling partners; for exam-
ple, 3-iodopyridine and 3-iodothiophene gave the desired products
3t–u in good yields. In order to further explore the generality of
this procedure, we turned our attention to further expand the
scope of the reaction to other ortho-hydroxycinnamate 1. For
example, the ortho-hydroxycinnamates 1b–f bearing methyl,
bromo, and chloro substituents were used and afforded the
Results and discussion
We began our study by examining the reaction of ethyl ortho-
hydroxy cinnamate 1a and phenyl iodide 2a as the model sub-
strates to optimize the formation of 4-arylcoumarin 3a under var-
ious conditions. Unexpectedly, the desired product of 3a was
obtained in 23% yield in the presence of 10 mol% Pd(OAc)2 as cat-
alyst and 2 equiv NaOAc as base in xylene at 100 °C for 12 h
(Table 1, entry 1). Initially, we screened various solvents, it was
found that the yield of 3a was slightly improved in dioxane or
MeCN (Table 1, entries 2,3). Moderate yields were obtained in
DMF or DMSO (Table 1, entries 4,5). To our delight, the desired pro-
duct 3a was obtained in 89% and 91% yields by using PEG-400 and
EtOH as solvents, respectively (Table 1, entry 6,7). However, only
moderate yields were obtained by using t-Amy-OH and AcOH as
solvents (Table 1, entries 8,9). Interestingly, we found that the sim-
ilar yield was obtained in H2O as a sole solvent (Table 1, entries 7 vs
10). Then, screening of other bases was attempted, the desired pro-
duct 3a was obtained in high yields by employing K3PO4 and
NaHCO3 as bases (Table 1, entries 11,12). The other bases such as
Na2CO3, Cs2CO3, K2CO3 and NaOH were less effective to promote
the transformation (Table 1, entries 13–16). Finally, the palladium
catalysts were investigated. Inferior results were given by Pd(C)
and Pd2(dba)3 (Table 1, entries 17,18). The other palladium salts
such as Pd(PPh3)4, PdCl2, PdCl2(PPh3)2 and Pd(OOCCF3)2 can also
promote the transformation in very high yields when the reaction
was carried out in similar condition reactions (Table 1, entries 19–
22). Finally, we found that PdCl2(CH3CN)2 was the best choice and
the desired product 3a was obtained in 95% yield (Table 1, entry
23). In addition, lowering the reaction temperature caused the
yield drastically descend (Table 1, entry 24). Attempts to lower
the catalyst loading from 10 to 5 mol% resulted in decreased yield