Synthesis of isocoumarins via Pd/C-mediated reactions of o-iodobenzoic acid with terminal alkynes

Article abstract of DOI:10.1021/jo050440e

The coupling reaction of o-iodobenzoic acid with terminal alkynes by using a catalyst system of 10% Pd/C-Et₃N-CuI-PPh₃ has been studied in a variety of solvents. 3-Substituted isocoumarins were formed in good yields and with good reg

Full text of DOI:10.1021/jo050440e

Synthesis of Isocoumarins via Pd/C-Mediated Reactions of
o-Iodobenzoic Acid with Terminal Alkynes†
Venkataraman Subramanian, Venkateswara Rao Batchu, Deepak Barange, and Manojit Pal*
Chemistry-Discovery Research, Dr. Reddy’s Laboratories Ltd., Bollaram Road,
Miyapur, Hyderabad 500049, India
manojitpal@drreddys.com
Received March 8, 2005
The coupling reaction of o-iodobenzoic acid with terminal alkynes by using a catalyst system of
10% Pd/C-Et3N-CuI-PPh3 has been studied in a variety of solvents. 3-Substituted isocoumarins
were formed in good yields and with good regioselectivity when the reaction was performed in
EtOH.
Introduction
The prevalence of isocoumarins1 in numerous natural
products that exhibit a wide range of pharmacological
activities, such as antifungal,2 antimicrobial,3 phytotoxic,4
and other effects, has led to a continued interest in the
practical synthesis of this class of lactones, especially
3-substituted isocoumarins. Considerable efforts have
been directed toward the synthesis of isocoumarins,5 via
either traditional or transition-metal-catalyzed reactions.
Among these, the Sonogashira type coupling followed by
the electrophilic6 or transition-metal-mediated cyclization
of the resulting alkynes possessing a carboxylate or an
equivalent group in proximity to the triple bond has
emerged as the most important process for the construc-
FIGURE 1. Construction of the isocoumarin ring.
tion of the isocoumarin ring (Figure 1). Thus, isocou-
marins have been prepared via cyclization of 2-(1-
alkynyl)benzoic acids/esters or amides in the presence of
a number of reagents: e.g., mineral acid5e or HBF4,5j
PdCl2(CH3CN)2,1f and silver,7a-c mercury,7d,e and copper
salts.8a Cyclization of o-phenylethynylbenzonitrile in the
presence of HgSO47d or Ru3(CO)128b leading to the 3-phen-
ylisocoumarin has also been reported. In an earlier effort
directed toward the one-pot synthesis of isocoumarins,5b
the reaction of o-iodobenzoic acids with copper acetylides
and terminal alkynes was investigated9 and found to
yield phthalides in most of the cases.9b,c The use of PdCl2-
† DRL publication number 476.
(1) (a) Mali, R. S.; Babu, K. N. J. Org. Chem. 1998, 63, 2488-2492
and references therein. (b) Barry, R. D. Chem. Rev. 1964, 64, 229-
260.
(2) Nozawa, K.; Yamada, M.; Tsuda, Y.; Kawai, K.; Nakajima, S.
Chem. Pharm. Bull. 1981, 29, 2491-2495.
(3) Hussain, M. T.; Rama, N. H.; Malik, A. Indian J. Chem., Sect. B
2001, 40, 372-376.
(6) For leading references, see: (a) Yao, T.; Larock, R. C. J. Org.
Chem. 2003, 68, 5936-5942. (b) Nagarajan, A.; Balasubramanian, T.
R. Indian J. Chem., Sect. B 1988, 27, 380. (c) Oliver, M. A.; Gandour,
R. D. J. Org. Chem. 1984, 49, 558-559. (d) Biagetti, M.; Bellina, F.;
Carpita, A.; Stabile, P.; Rossi, R. Tetrahedron 2002, 58, 5023-5038.
(e) Rossi, R.; Carpita, A.; Bellina, F.; Stabile, P.; Mannina, L.
Tetrahedron 2003, 59, 2067-2081. (f) Peng, A.-Y.; Ding. Y.-X. Org. Lett.
2004, 6, 1119-1121.
(4) Sato, H.; Konoma, K.; Sakamura, S. Agric. Biol. Chem. 1981,
45, 1675-1679.
(5) (a) Korte, D. E.; Hegedus, L. S.; Wirth, R. K. J. Org. Chem. 1977,
42, 1329-1336. (b) Batu, G.; Stevenson, R. J. Org. Chem. 1980, 45,
1532-1534. (c) Larock, R. C.; Varaprath, S.; Lau, H. H.; Fellows, C.
A. J. Am. Chem. Soc. 1984, 106, 5274-5284. (d) Izumi, T.; Nishimoto,
Y.; Kohei, K.; Kasahara, A. J. Heterocycl. Chem. 1990, 27, 1419-1424.
(e) Villemin, D.; Goussu, D. Heterocycles 1989, 29, 1255-1261. (f)
Sashida, H.; Kawamukai, A. Synthesis 1999, 1145-1148. (g) Izumi,
T.; Morishita, N. J. Heterocycl. Chem. 1994, 31, 145-152. (h) Napoli-
tano, E. Org. Prep. Proced. Int. 1997, 29, 631. (i) Rossi, R.; Bellina, F.;
Biagetti, M.; Catanese, A.; Mannina, L. Tetrahedron Lett. 2000, 41,
5281-5286. (j) Tovar, J. D.; Swager, T. M. J. Org. Chem. 1999, 64, 4;
6499-6504. (k) Larock, R. C.; Doty, M. J.; Han, X. J. Org. Chem. 1999,
64, 4; 8770-8779.
(7) (a) Bellina, F.; Ciucci, D.; Vergamini, P.; Rossi, R. Tetrahedron
2000, 56, 2533-2545. (b) Ogawa, Y.; Maruno, M.; Wakamatsu, T.
Synlett 1995, 871-872. (c) Ogawa, Y.; Maruno, M.; Wakamatsu, T.
Heterocycles 1995, 41, 2587-2599. (d) Sakamoto, T.; An-naka, M.;
Kondo, Y.; Yamanaka, H. Chem. Pharm. Bull. 1986, 34, 2754-2759.
(e) Nagarajan, A.; Balasubramanian, T. R. Indian J. Chem., Sect. B
1987, 26, 917-919.
(8) (a) Peng, A.-Y.; Ding, Y.-X. J. Am. Chem. Soc. 2003, 125, 15006-
15007. (b) Menashe, N.; Shvo, Y. Heterocycles 1993, 35, 611-613.
10.1021/jo050440e CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/17/2005
4778
J. Org. Chem. 2005, 70, 4778-4783

Synthesis of Isocoumarins
(PPh3)2-Et3N-CuI as a catalyst system also exhibited
greater selectivity for phthalides for a variety of terminal
acetylenes.10 However, in a subsequent study Cheng and
Liao have shown that isocoumarins could be obtained as
major products when o-iodobenzoic acid was reacted with
terminal alkynes in the presence of Pd(PPh3)4, Et3N, and
a stoichiometric amount of ZnCl2.11 The use of ZnCl2 in
place of CuI was found to be responsible for the predomi-
nant formation of isocoumarins over phthalides. In
contrast, we have noted that o-iodobenzoic acid reacts
smoothly with terminal alkynes in the presence of CuI
and Pd/C-Et3N-PPh3 as a catalyst system in ethanol,
affording the corresponding isocoumarins in good yields.
While there have been numerous studies on Pd/C-
mediated coupling of aryl halides with terminal acety-
lenes,12 to the best of our knowledge the use of Pd/C for
the synthesis of isocoumarins or phthalides has never
been reported thus far. In this study, we first report the
palladium (on charcoal)-copper-mediated synthesis of
diversified 3-substituted isocoumarins from o-iodobenzoic
acid and terminal alkynes.
FIGURE 2. Pd/C-mediated synthesis of 2-substituted benzo-
[b]furans and indoles.14
TABLE 1. Effect of Reaction Conditions on the
Palladium-Catalyzed Coupling Reaction of o-Iodobenzoic
Acid with 2-Methyl-3-butyn-2-ola
yield (%)b
entry
Pd catalyst
base; solvent
(IIIa:IVa)c
1
2
3
4
5
6
7
8d
10% Pd/C-PPh3
10% Pd/C-PPh3
10% Pd/C-PPh3
10% Pd/C-PPh3
10% Pd/C-PPh3
Pd(PPh3)2Cl2
Et3N; DMF
25 (4:1)
Et3N; EtOH
Et3N; iPrOH
Et3N, tBuOH
Et3N; 1,4-dioxane
Et3N; EtOH
Et3N; EtOH
Et3N; EtOH
75 (IVa only)
92 (1:4)
80 (1:4)
85 (9:1)
82 (9:1)
Results and Discussion
Pd(PPh3)4
90 (1:9)
10% Pd/C-PPh3
55 (IVa only)
In pursuance of our research under the new drug
development program,13a,b we became interested in the
synthesis of oxygen-containing heterocycles,13a,c-f espe-
cially 3-substituted isocoumarin, of potential pharmaco-
logical interest. Our interest in the 3-substituted isocou-
marin originated from the fact that the angiogenesis
inhibitor NM-3 (2-(8-hydroxy-6-methoxy-1-oxo-1H-2-ben-
zopyran-3-yl)propionic acid),13g which belongs to this
class, is presently undergoing Phase 1 clinical trials.
During our Pd/C-mediated synthesis14 of 2-substituted
benzo[b]furans14a and indoles14b (Figure 2), we were
intrigued with the possibility of using Pd/C as a cheap
and alternative source of palladium catalyst for the
coupling of 2- halobenzoic acid (I) with terminal alkynes.
Accordingly, we carried out a number of experiments
using 10% Pd/C-Et3N-CuI-PPh3 as a catalyst system
in a variety of solvents (Table 1). At the beginning of our
a Reaction conditions: I (1.0 equiv), IIa (2.0 equiv), 10% Pd/C
or other Pd catalyst (0.03 equiv), PPh3 (0.12 equiv), CuI (0.06
equiv), Et3N (5 equiv) in a solvent at 80 °C for 16 h under N2.
b Isolated overall yield of IIIa + IVa. c The ratio was determined
by 1H NMR analysis. d The reaction was carried out for 24 h.
study the reaction was carried out in DMF (entry 1, Table
1) where phthalide (IIIa) was isolated as the major
product, although in low yield. Notably, DMF was the
solvent of choice in the earlier synthesis of phthalides
and isocoumarins in the presence of palladium com-
plexes.15 Switching over from DMF to EtOH, we observed
a dramatic change in the product distribution, and the
corresponding isocoumarin (IVa) was isolated as the sole
product in this case (entry 2, Table 1). This observation
prompted us to examine the use of other alcohols: e.g.,
t
iPrOH and BuOH (entries 3 and 4, Table 1). While the
overall yields of products were improved marginally, a
considerable amount of phthalide was detected as a side
(9) In an initial report Castro and co-workers claimed that the
reaction of o-iodobenzoic acid with cuprous phenylacetylide led to the
formation of 3-phenylisocoumarin; see: (a) Castro, C. E.; Stephens, R.
D. J. Org. Chem. 1963, 28, 2163. (b) Castro, C. E.; Stephens, R. D. J.
Org. Chem. 1963, 28, 3313-3315. In a later paper the structure of
3-phenylisocoumarin was corrected to 3-benzylidenephthalide by the
same group; see: (c) Castro, C. E.; Gaughan, E. J.; Owsley, D. C. J.
Org. Chem. 1966, 31, 4071-4078. (d) Okuro, K.; Furuune, M.; Enna,
M.; Miura, M.; Nomura, M. J. Org. Chem. 1993, 58, 4716-4721. For
the synthesis of 3-phenylisocoumarin under microwave irradiation,
see: (e) Wang, J.-X.; Liu, Z.; Hu, Y.; Wei, B. J. Chem. Res., Synop.
2000, 11, 536-537. (f) Wang, J.-X.; Liu, Z.; Hu, Y.; Wei, B.; Kang, L.
Synth. Commun. 2002, 32, 1937-1946.
(13) (a) Pal, M.; Veeramaneni, V. R.; Nagabelli, M.; Kalleda, S. R.;
Misra, P.; Casturi, S. R.; Yeleswarapu, K. R. Bioorg. Med. Chem. Lett.
2003, 13, 1639-1643. (b) Pal, M.; Madan, M.; Srinivas, P.; Pattabira-
man, V. R.; Kalleda, S. R.; Akhila, V.; Ramesh, M.; Rao Mamidi, N. V.
S.; Casturi, S. R.; Malde, A.; Gopalakrishnan, B.; Yeleswarapu, K. R.
J. Med. Chem. 2003, 46, 3975-3984. (c) For 3-alkynyl flavones, see:
Pal, M.; Subramanian, V.; Parasuraman, K.; Yeleswarapu, K. R.
Tetrahedron 2003, 59, 9563-9570. (d) For 3-enynyl thioflavones/
flavones, see: Pal, M.; Parasuraman, K.; Subramanian, V.; Dakarapu,
R.; Yeleswarapu, K. R. Tetrahedron Lett. 2004, 45, 2305-2309. (e) For
the synthesis of 3,4-diaryl maleic anhydrides, see: Pattabiraman, V.
R.; Padakanti, P. S.; Veeramaneni, V. R.; Pal, M.; Yeleswarapu, K. R.
Synlett 2002, 947-951. (f) For the synthesis of 3,4-diaryl furanones,
see: Padakanti, S.; Veeramaneni, V. R.; Pattabiraman, V. R.; Pal, M.;
Yeleswarapu, K. R. Tetrahedron Lett. 2002, 43, 8715-8719. (g) Oikawa,
T.; Sasaki, M.; Inose, M.; Shimamura, M.; Kuboki, H.; Hirano, S.-I.;
Kumagai, H.; Ishizuka, M.; Takeuchi, T. Anticancer Res. 1997, 17,
1881-1886.
(14) (a) Pal, M.; Subramanian, V.; Yeleswarapu, K. R. Tetrahedron
Lett. 2003, 44, 8221-8225. (b) Pal, M.; Subramanian, V.; Batchu, V.
R.; Dager, I. Synlett 2004, 1965-1969.
(15) As noted by Villemin and Goussu initially, in Et3N alone the
reaction mixture containing o-iodobenzoic acid became heterogeneous
and no reaction was observed. A homogeneous mixture was obtained
with dichloromethane, but no reaction took place.5e This problem was
overcome by Kundu and Pal by using DMF as a cosolvent, where the
reaction proceeded smoothly to yield phthalides as major products.10
(10) (a) Kundu, N. G.; Pal, M.; Nandi, B. J. Chem. Soc., Perkin
Trans. 1 1998, 561-568. (b) Kundu, N. G.; Pal, M. J. Chem. Soc., Chem.
Commun. 1993, 86-88.
(11) Liao-H.-Y.; Cheng, C.-H. J. Org. Chem. 1995, 60, 3711-3716.
(12) (a) Novak, Z.; Szabo, A.; Repasi, J.; Kotschy, A. J. Org. Chem.
2003, 68, 3327-3329. (b) For the use of Pd(OH)2/C as catalyst, see:
Mori, Y.; Seki, M. J. Org. Chem. 2003, 68, 1571-1574. (c) Heidenreich,
R. G.; Ko¨hler, K.; Krauter, J. G. E.; Pietsch, J. Synlett 2002, 1118-
1122. (d) Felpin, F.-X.; Vo-Thanh, G.; Villie´ras, J.; Lebreton, J.
Tetrahedron: Asymmetry 2001, 12, 1121-1124. (e) Bates, R. W.;
Boonsombat, J. J. Chem. Soc., Perkin Trans. 1 2001, 654-657. (f)
Bleicher, L. S.; Cosford, N. D. P.; Herbaut, A.; McCallum, J. S.;
McDonald, I. A. J. Org. Chem. 1998, 63, 1109-1118. (g) Bleicher, L.;
Cosford, N. D. P. Synlett 1995, 1115-1116. (h) Potts, K. T.; Horwitz,
C. P.; Fessak, A.; Keshavarz, K. M.; Nash, K. E.; Toscano, P. J. J. Am.
Chem. Soc. 1993, 115, 10444-10445. (i) De la Rosa, M. A.; Velarde,
E.; Guzman, A. Synth. Commun. 1990, 20, 2059-2064.
J. Org. Chem, Vol. 70, No. 12, 2005 4779

Subramanian et al.
product in such cases. The use of 1,4-dioxane afforded
IIIa as a major product under the present Pd/C-Cu
catalysis (entry 5, Table 1). The use of other catalysts,
e.g. Pd(PPh3)2Cl2 and Pd(PPh3)4 (originally used for the
synthesis of phthalides and isocoumarins, respectively),
were examined (entries 6 and 7, Table 1). Remarkably,
Pd(PPh3)4 showed selectivity for isocoumarin, in contrast
to a Pd(PPh3)2Cl2-catalyzed reaction where phthalide was
isolated as the major product. Nevertheless, unlike the
case for Pd/C, the isocoumarin formed in this case was
contaminated with the corresponding phthalide (entry 2
vs 7, Table 1). The role of PPh3 was also evaluated, in
the absence of which no desired product was detected
under the reaction conditions studied. We then examined
the use of LiCl (1-5 equiv) and its effect on the reaction
conditions as described above (cf. entry 2, Table 1). The
use of LiCl was found to be beneficial for the synthesis
of 3,4-disubstituted isocoumarins from 2-iodobenzoate
ester and internal alkynes.5k Although the reaction was
found to be faster (8 h vs 16 h of entry 2, Table 1) when
an excess of LiCl (5 equiv) was used in the present
synthesis of IVa, no significant effect on product ratio
(i.e. IIIa/IVa) or yield was observed. Thus, after a series
of experiments, 10% Pd/C in ethanol proved to be the
most effective combination of catalyst and solvent for the
synthesis of isocoumarins. This Pd/C-mediated coupling
reaction in ethanol was carried out for 16 h, as an
increase in reaction time to 24 h or more resulted in the
ethanolysis of the product formed under the prolonged
heating conditions, thereby decreasing the product yield
(entry 8, Table 1).
We were delighted to observe the formation of 3-sub-
stituted isocoumarin as a sole product under certain
reaction conditions and, therefore, decided to test the
reaction conditions with other terminal alkynes. Using
the optimized protocol as detailed above (entry 2, Table
1), several 3-substituted isocoumarins were prepared in
ethanol. Thus, when o-iodobenzoic acid (I) was treated
with 2 equiv of terminal alkyne (II; R ) aryl, alkyl,
hydroxyalkyl, etc.)16a in EtOH in the presence of 10%
Pd/C (0.03 equiv), PPh3 (0.12 equiv), CuI (0.06 equiv),
and Et3N (5 equiv) under a nitrogen atmosphere iso-
coumarins16d (IV) were obtained in good yields.
By use of this Pd/C-catalyzed tandem coupling-cy-
clization process a variety of commercially available
terminal acetylenes were reacted with o-iodobenzoic
acids,16b,c and the yields of isolated products (IV and V)
after purifying by column chromatography have been
presented in Tables 2 and 3. Various functional groups
(including alkyl, hydroxyl, phenyl, etc.) present in acety-
lenic compounds (II) employed so far were well tolerated
during the course of the reaction (Table 2). The process
was found to be quite general for the preparation of
3-substituted isocoumarins. While a primary or second-
ary alcohol present in the terminal alkynes was readily
accommodated (entries 2-4, Table 2), the presence of a
long-chain alkyl group did not affect the coupling-
cyclization process (entries 5-7, Table 2). Of particular
note, however, are two examples (entries 8 and 9, Table
2) where in one case a significant amount of side product,
i.e., 3-phenyl-4-(phenylethynyl)isocoumarin, was isolated
along with the desired isocoumarin when phenylacety-
lene was used (entry 8, Table 2). In another case an
alkyne bearing a SiMe3 group afforded the five-membered
ring product exclusively after subsequent desilylation of
the resulting phthalide in one pot (entry 9, Table 2).
2-((Trimethylsilanyl)ethynyl)benzoic acid (IIIbb) was
isolated as other major product in this case. In general,
the coupling-cyclization reaction was carried out using
2 equiv of terminal alkynes (II). The use of a lesser
amount, i.e. 1.5 equiv, of II also afforded IV, in good
yields except in the case of IIa,e,f,i, where the corre-
sponding products were isolated in low yields. Slow
evaporation of II (due to the volatile nature of these
alkynes) could be the reason for such observations. We
next examined the possibility of using diversified o-
iodobenzoic acids in this coupling-cyclization reaction.
As expected, the substitutions on the benzene ring did
not effect the reaction and, thus, isocoumarins bearing
electron-donating and/or electron-withdrawing groups at
the 5- and/or 7-positions of the aromatic ring were
prepared in good yields (Table 3). The generality of this
process was demonstrated further by synthesizing a 6,7-
disubstituted isocoumarin in 70% yield (entry 7, Table
3).
The goal of this research was to develop an efficient
and cheaper method for the synthesis of an isocoumarin
library under mild and more environmental friendly
conditions. We have addressed these issues partially by
replacing the expensive palladium catalysts and solvent
with less expensive 10% Pd/C and ethanol. Taking into
consideration the scale-up potential of this process, we
next examined the possibility of preparing isocoumarins
by using this methodology in water. However, we failed
to isolate the desired product when the reaction of I with
terminal alkyne IIa was carried out in water instead of
EtOH under the same reaction conditions as described
above (entry 2, Table 1). The use of 2-aminoethanol in
place of triethylamine afforded no isocoumarin or phtha-
lide but a different coupled product, i.e., 2-((2-hydroxy-
(16) (a) o-Iodobenzoic acid and all the terminal alkynes used are
commercially available. (b) 5-Nitro-2-iodobenzoic acid was prepared
via nitration of o-iodobenzoic acid according to the procedure described
in the literature; see: Goldstein, H.; Grampoloff, A. V. Helv. Chim.
Acta 1930, 13, 310-314. (c) 2-Iodo-3-methoxybenzoic acid was prepared
via diazotization of 2-amino-3-methoxybenzoic acid according to the
procedure described in the literature; see: Kenner, J.; Turner, H. A.
J. Chem. Soc. 1928, 2340-2343. For the preparation of 2-iodo-4,5-
dimethoxybenzoic acid from 2-amino-4,5-domethoxybenzoic acid, see:
Kundu, N. G.; Khan, M. W. Tetrahedron 2000, 56, 4777-4792. (d) The
spectral and other characterization data for selected compounds are
as follows. IVa: white solid; mp 66-68 °C (lit10a mp 70-72 °C); 1H
NMR (200 MHz, CDCl3) δ 8.27 (d, J ) 8.0 Hz, 1H), 7.67 (t, J ) 6.5 Hz,
1H), 7.49-7.26 (m, 2H), 6.62 (s, 1H, CHdC), 1.99 (bs, 1H, OH), 1.60
(s, 6H, 2CH3); IR (KBr, cm-1): 3381 (bs, OH), 1733 (CdO); mass (m/z)
204 (M+, 50%), 189 (M+ - 15, 100%); 13C NMR (50 MHz, CDCl3) 162.4
(CdO), 162.3, 137.2, 134.6, 129.2, 127.7, 125.7, 119.8, 99.7 (CHdC),
70.5 (CMe2OH), 30.9 (CH3), 28.1 (CH3); UV (MeOH, nm) 323.0, 272.0,
263.0, 239.0, 228.0; HPLC 97.3%, column Inertsil ODS 3V (150 × 4.6)
mm, mobile phase A 0.01 M KH2PO4, mobile phase B CH3CN, gradient
(T, °C/% B) 0/20, 3/20, 15/80, 20/80, 21/20, 22/20, flow rate 1.5 mL/
min, UV 230 nm, retention time 8.4 min. IVd: white solid; mp 64-66
°C; 1H NMR (200 MHz, CDCl3) δ 8.25 (d, J ) 8.1 Hz, 1H), 7.69-7.65
(m, 1H), 7.48-7.44 (m, 1H), 7.35 (d, J ) 8.1 Hz, 1H), 6.30 (s, 1H, CHd
C), 3.75 (t, J ) 6.2 Hz, 2H, CH2), 2.67 (t, J ) 7.5 Hz, 2H, CH2), 2.02-
1.95 (m, 2H, CH2), 1.56 (1H, D2O exchangeable, OH); IR (KBr, cm-1
)
3435 (bs, OH), 1738 (CdO), 1656; mass (m/z) 205 (M+ + 1, 100%), 187
(M+ - 18, 50%); 13C NMR (50 MHz, CDCl3) δ 162.2 (CdO), 157.4, 137.5,
134.7, 129.4, 127.6, 125.0, 120.0, 103.3 (CHdC), 61.5 (CH2), 29.9 (CH2),
15.7 (CH2); UV (MeOH, nm) 326.0, 273.0, 264.0, 240.0, 229.0; HPLC
99.3%, Inertsil ODS 3V (150 × 4.6 mm) mobile phase A 0.01 M KH2-
PO4, mobile phase B CH3CN, gradient (T, °C/% B) 0/25, 3/25, 12/80,
20/80, 21/25, 22/25; 1.5 mL/min, UV 230 nm, retention time 6.33 min.
(e) For a similar coupling reaction of o-iodobenzoic acid with primary
amine, e.g. L-valine, see: Ma, D.; Zhang, Y.; Yao, J.; Wu, S.; Tao, F. J.
Am. Chem. Soc. 1998, 120, 12459-12467.
4780 J. Org. Chem., Vol. 70, No. 12, 2005

Synthesis of Isocoumarins
TABLE 2. Pd/C-Catalyzed Synthesis of 3-Substituted
Isocoumarinsa
ethyl)amino)benzoic acid, within 2 h in 72% yield,
presumably due to the increased nucleophilicity of the
amine (see later for mechanistic discussion).16e The use
of an inorganic base such as K2CO3 also failed to afford
the desired product in aqueous media.
Mechanistically,10,11 this reaction seems to proceed via
in situ generation of 2-(1-alkynyl)benzoic acid in the
presence of 10% Pd/C, PPh3, and CuI, which in turn
undergoes 6-endo-dig cyclization17 in a regioselective
manner to give the six-membered isocoumarin. This was
further supported by the isolation of IIIbb (entry 9, Table
2), where apparently the stability of the cationic inter-
mediate and the steric bulk of the SiMe3 group affected
the cyclization of the coupled product (see later for
mechanistic discussion). Recent studies12a,18 have shown
that the minor portion of the bound palladium (Pd/C)
released into the solution is the actual catalytic species,
indicating that the catalytic cycle works in solution rather
than on the surface and the active species is a dissolved
Pd-PPh3 complex.19 On the other hand, according to
Amatore and Jutand,20a,b halide ions play a specific role
in generating an anionic species such as [L2Pd0Cl]- from
Pd(PPh3)2X2, which is thought to be the active palladium
species in the cross-coupling reactions.20c On the basis
of this information a plausible mechanism is proposed
in Scheme 1.
The anionic species [(PPh3)2Pd0I]- (X) generated in situ
from Pd/C and PPh3 and then in the presence of CuI/
terminal alkynes thus catalyzes the coupling of o-iodo-
benzoic acid with copper(I) acetylides (generated in situ
from terminal alkynes) via intermediates20d Y and Z,
leading to the formation of 2-(1-alkynyl)benzoic acid,
which subsequently undergoes cyclization to yield the
corresponding isocoumarin.20e The enhanced rate of reac-
tion observed in the presence of excess LiCl can be
explained by coordination of the chloride with palladium
in the catalyst activation step to form a chloride-ligated
(17) (a) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734-
736. (b) Baldwin, J. E.; Cutting, J.; Dupont, W.; Kruse, L.; Silberman,
L.; Thomas, R. C. J. Chem. Soc., Chem. Commun. 1976, 736-738.
(18) Mori, Y.; Seki, M. J. Org. Chem. 2003, 68, 1571-1574.
(19) De la Rosa, M. A.; Velarde, E.; Guzman, A. Synth. Commun.
1990, 20, 2059-2064.
(20) (a) Amatore, C.; Jutand, A. Acc. Chem. Res. 2000, 33, 314-
321. (b) Amatore, C.; Jutand, A.; Khalil, F.; M’Barki, M. A.; Mottier,
L. Organometallics 1993, 12, 3168. (c) The use of 4 equiv of PPh3 was
necessary to achieve the best catalytic activity of the Pd/C used. In
addition to the other donors, i.e., Et3N and EtOH, the excess of PPh3
presumably interacted with the bound palladium that helped in
releasing the portion of palladium into the solution. (d) The σ-coordi-
nated benzoic acid in intermediates Y and Z may alternatively be
represented by the structures shown below, taking into consideration
of the fact that either of the carboxylic acid oxygens would form a
perfect five-membered chelate, most likely displacing the extra iodide,
as the binding strength of the chelate would be much greater than
the Pd-I bond strength (we thank one of the reviewers for bringing
this to our attention). In ethanol the intermediate Z, however, would
afford 2-(phenylethynyl)benzoic ester as a main or side product in
addition to the desired isocoumarins (IV) as shown below. Since this
ester was never isolated even in trace quantity from the reaction
mixture, the intermediacy of the chelated benzoic acid in Y and Z would
seem to be unlikely. (e) Grosshenny, V.; Romero, F. M.; Ziessel, R. J.
Org. Chem. 1997, 62, 1491-1500.
a All reactions were carried out by using I (1.0 equiv), II (2.0
equiv), 1/4/2 ratio of 10%Pd/C, PPh3, and CuI, Et3N (5 equiv) in
EtOH for 16 h. b Identified by 1H NMR, 13C NMR, IR, mass. c A
side product, i.e., 3-phenyl-4-(phenylethynyl)isocoumarin, was
isolated in 25% yield. d Overall yield.
zerovalent palladium species that subsequently catalyzes
the coupling-cyclization reaction. In the presence of
2-aminoethanol the 18-electron complex Y (generated by
the oxidative addition of the aryl iodide to the zerovalent
anion X and in which the iodide ion, borne by the Pd(0),
remains attached to the Pd(II) center) undergoes elimi-
nation of iodide ion to yield a neutral pentacoordinated
complex. This complex affords 2-((2-hydroxyethyl)amino)-
benzoic acid via reductive elimination, followed by the
regeneration of the active palladium species X. For the
J. Org. Chem, Vol. 70, No. 12, 2005 4781

Subramanian et al.
TABLE 3. Pd/C-Catalyzed Synthesis of Functionalized
3-Substituted Isocoumarinsa
SCHEME 1. Plausible Mechanism for the
Formation of Isocoumarin
SCHEME 2. Intramolecular Cyclization of
2-(1-Alkynyl)benzoic Acid
in EtOH and 1,4-dioxane separately at 80 °C for 2 h. The
reaction in ethanol gave the corresponding isocoumarin
in 70% yield, whereas a 1:1 mixture of phthalide and
isocoumarin was isolated in 60% overall yield in 1,4-
dioxane, indicating that the combined effect of CuI and
PPh3 was likely responsible for the preferred 6-endo-dig
cyclization in ethanol. Notably, despite its well-known
role in the formation of phthalides, Cu(I) salts are also
known to catalyze the intramolecular cyclization of 2-(1-
alkynyl)benzoic acid or its derivative to the six-membered
lactone ring.5b,9d,f,21b,c The possibility of a Pd(II)-catalyzed
cyclization of 2-(1-alkynyl)benzoic acid that usually af-
fords isocoumarin as a predominant product5f can be
ruled out, because the existence of such a species is
unlikely under the reaction conditions studied. The
participation of the anionic species Y, which contains a
Pd(II) center and could catalyze the cyclization step to
afford a 4-aryl-substituted isocoumarin, was also ruled
out, because the corresponding product was not detected
in the reaction mixture.
a For reaction conditions see footnote a in Table 2. b Identified
1
by H NMR, 13C NMR, IR, mass. c For IIj, R ) CH2CH(OH)CH3.
intramolecular cyclization of 2-(1-alkynyl)benzoic acid,
the “6-endo-dig” or the “5-exo-dig” closure is allowed by
Baldwin’s rule. However, reasons for the observed regio-
selectivity associated with the use of 10% Pd/C-CuI-
PPh3 as a catalyst system are not yet clear at this stage.
The direction of favored cyclization seems to be influenced
greatly by the nature of the catalysts as well as solvent
used in the present case. The 6-endo-dig cyclization was
favored over 5-exo-dig closure in ethanol, possibly due
to the higher stability of the six-membered cationic
species in a polar protic solvent leading to the formation
of endocyclic product (Scheme 2). However, to gain
further evidence regarding the role of solvent in the
present coupling-cyclization reaction 2-(phenylethynyl)-
benzoic acid21a (known to be an intermediate in the
palladium-copper-catalyzed coupling-cyclization of phen-
ylacetylene with I)10a was treated with CuI-Et3N-PPh3
The structures of all 3-substituted isocoumarins were
1
confirmed by satisfactory spectroscopic (MS, IR, and H
NMR) data. The isocoumarins can be differentiated from
the corresponding phthalides (that are usually found to
be side products in the other palladium-mediated syn-
(21) (a) This compound was prepared according to the procedure
described in ref 10a. (b) For Cu(I)-catalyzed intramolecular cyclization
of o-ethynylphosphonic acid monoesters to phosphaisocoumarins, see
ref 8a and: (c) Ivanchikova, I. D.; Usubalieva, G. E.; Schastnev, P. V.;
Moroz, A. A.; Shvartsberg, M. S. Izv. Akad. Nauk, Ser. Khim. 1992, 9,
2138-2146; Chem. Abstr. 1993, 118, 123864.
4782 J. Org. Chem., Vol. 70, No. 12, 2005

Synthesis of Isocoumarins
theses of isocoumarins) on the basis of several spectro-
Pentylisocoumarin (IVf) showed significant antibacterial
activity comparable to the standard antibiotics.22c Further
application of this process to generate isocoumarin-based
chemical libraries of potential pharmacological interest
is currently underway.
1
scopic data, especially H NMR and IR spectra. On the
basis of the reported10a comparison between the six- and
five-membered lactone rings of isocoumarins (νmax(CdO)
1710-1750 cm-1 in IR and δH(vinylic) 6.3-7.0 in 1H NMR
spectra) and 3-ylidenephthalides (νmax(CdO) 1770-1800
1
cm-1 in IR and δH(vinylic) 5.0-7.0 in H NMR spectra),
Experimental Section
respectively, we found that our observations were in full
agreement with the spectral data reported for iso-
coumarins.
Typical Procedure for the Preparation of IVe. A
mixture of I (0.3 g, 1.21 mmol), 10% Pd/C (35 mg, 0.033 mmol),
PPh3 (38 mg, 0.14 mmol), CuI (14 mg, 0.07 mmol), and Et3N
(0.61 g, 0.84 mL, 6.05 mmol) in EtOH (15 mL) was stirred at
25 °C for 30 min under nitrogen. The acetylenic compound IIe
(0.199 g, 0.28 mL, 2.42 mmol) was added slowly with stirring.
The mixture was then stirred at 80 °C for 16 h, cooled to room
temperature, diluted with EtOAc (50 mL), and filtered through
Celite. The filtrate was collected and concentrated. The residue
was purified by column chromatography (petroleum ether-
EtOAc) to afford the desired product (0.146 g, 60% yield).
We have shown that the Pd/C-Et3N-PPh3-CuI cata-
lyst system allows smooth and effective coupling of
2-iodobenzoic acid with terminal alkynes. A detailed
study on the methodology along with its limitations has
also been described. In ethanol this catalyst system can
be utilized for a practical, one-pot synthesis of isocou-
marins. In contrast to the PdCl2(PPh3)2-Et3N-CuI sys-
tem in DMF this process exhibited greater selectivity for
isocoumarins. Moreover, unlike the Pd(PPh3)4-Et3N-
ZnCl2 system the present methodology does not involve
the use of expensive as well as air-sensitive palladium-
(0) catalyst and a substantial quantity of ZnCl2. Since
many terminal alkynes are commercially available or
synthetically accessible, the described methodology, be-
cause of its operational simplicity, holds promise espe-
cially in the synthesis of isocoumarin-based molecules for
thorough SAR studies in a medicinal chemistry setting.
Of the several isocoumarins synthesized, 3-butylisocou-
marin (IVe) is a precursor for a naturally occurring
isocoumarin, i.e., artemidin,5b and compounds of biologi-
cal interest.22a,b Moreover, it can be converted to the
corresponding 3-substituted isoquinolone easily.7d 3-
Acknowledgment. We thank Dr. A. Venkateswarlu,
Dr. R. Rajagopalan, and Prof. J. Iqbal for their encour-
agement and the analytical group for spectral support.
Supporting Information Available: Text giving experi-
mental procedures and spectral data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO050440E
(22) (a) Shafiq, Z.; Arfan, M.; Rama, N. H.; Hussain, M.; Hameed,
S. Indian J. Chem., Sect. B 2003, 42, 1523-1526. (b) The 3-butyl side
chain has been found to be a prerequisite for high antifungal activity
of several naturally occurring isocoumarins; see: Engelmeier, D.;
Hadacek, F.; Hofer, O.; Lutz-Kutschera, G.; Nagl, M.; Wurz, G.; Greger,
H. J. Nat. Prod. 2004, 67, 19-25. (c) Rama, N. H.; Iqbal, R.; Zamani,
K.; Saeed, A.; Iqbal, M. Z.; Choudhary, M. Indian J. Chem., Sect. B
1998, 37, 365-369.
J. Org. Chem, Vol. 70, No. 12, 2005 4783