Novel PdII-mediated cascade carboxylative annulation to construct benzo[b]furan-3-carboxylic acids

Article abstract of DOI:10.1021/ol050876g

(Chemical Equation Presented) Benzo[b]furan-3-carboxylic acid (2) was generated from 1 by forming three new bonds in one step via a Pd ^II-mediated cascade carboxylative annulation. The proposed mechanism was supported by the observation of an u

Full text of DOI:10.1021/ol050876g

ORGANIC
LETTERS
Novel PdII-Mediated Cascade
Carboxylative Annulation to Construct
Benzo[b]furan-3-carboxylic Acids
2005
Vol. 7, No. 13
2707-2709
Yun Liao,*,† Jennifer Smith, Reza Fathi,*,† and Zhen Yang‡
ViVoQuest, Inc., 711 ExecutiVe BlVd, Valley Cottage, New York 10989, and
Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of
Education, Department of Chemical Biology, College of Chemistry, Peking UniVersity,
Beijing, P. R. China 100871
y.liao@ViVoquest.com; r.fathi@ViVoquest.com
Received April 20, 2005
ABSTRACT
Benzo[b]furan-3-carboxylic acid (2) was generated from 1 by forming three new bonds in one step via a PdII-mediated cascade carboxylative
annulation. The proposed mechanism was supported by the observation of an unusual acetylation of 1 as a side reaction together with an
18O-labeling study.
Recently, the PdII-catalyzed/mediated cascade carbonylative
annulation of the o-hydroxyarylacetylenes (1) has proved a
highly efficient method for rapidly generating diverse benzo-
[b]furan-3-carboxylic esters (3) (a, Scheme 1).1 However,
edented, although Pd-promoted carboxylation of unsaturated
carbon-carbon bonds is well documented especially via
Reppe carbonylation.2
From a drug discovery perspective, synthesis of benzo-
furan-based carboxylic acids could be more interesting
because of their increased solubility in aqueous media and
the potential enhancement of ionic interactions with basic
residues in their association with biological receptors.3 So
far, there is only one example of a direct method for making
2 from 4 (c, Scheme 1, R1 ) H, R2 ) Me, 62% yield).4
However, this method employed a strong reagent, TBAF,
for success. Although molecules such as 2 can also be
obtained by the hydrolysis of 3, a strong base is necessary
since 3 is already stabilized by the electron-rich benzofuran
ring. Therefore, direct and mild methods to make structurally
diverse molecules such as 2 still need to be developed.
Scheme 1
(2) (a) Kiss, G. Chem. ReV. 2001, 101, 3435 and references therein. (b)
Ohashi, S.; Sahaguchi, S.; Ishii, Y. Chem. Commun. 2005, 486. (c) Fujiwara,
Y.; Kawauchi, T.; Taniguchi, H. J. Chem. Soc., Chem. Commun. 1980,
220. (d) Jintoku, T.; Taniguchi, T.; Fujiwara, Y. Chem. Lett. 1987, 1159.
(e) Jia, C.; Kitamura, T.; Fujiwara, Y. Acc. Chem. Res. 2001, 34, 633. (f)
Fujiwara, Y.; Takaki, K.; Taniguchi, Y. Synlett 1996, 591. (g) Urata, H.;
Fujita, A.; Fuchikami, T. Tetrahedron Lett. 1988, 29 (35), 4435.
(3) Remy, D. C.; Britcher, S. F.; King, S. W. J. Med. Chem. 1983, 26,
974.
similar metal-mediated transformations leading to benzo[b]-
furan-3-carboxylic acids (2) (b, Scheme 1) remain unprec-
† VivoQuest, Inc.
‡ Peking University.
(1) (a) Lu¨tjens, H.; Scammells, P. J. Tetrahedron Lett. 1998, 39, 6581.
(b) Lu¨tjens, H.; Scammells, P. J. Synett 1999, 7, 1079. (c) Nan, Y.; Miao,
H.; Yang, Z. Org. Lett. 2000, 2, 297. (d) Liao, Y.; Reitman, M.; Zhang,
Y.; Fathi R.; Yang, Z. Org. Lett. 2002, 4, 2607.
(4) Aoyma, T.; Shioiri, T. Synlett 1997, 10, 1163.
10.1021/ol050876g CCC: $30.25
© 2005 American Chemical Society
Published on Web 05/24/2005

Our former endeavors1c,d in generating diverse benzo[b]-
furan-3-carboxylic esters (3) (a, Scheme 1) using Pd
chemistry encouraged us to further explore this cascade PdII
chemistry for constructing the benzofuran-3-carboxylic acids
(2) (b, Scheme 1) directly from the o-hydroxyarylacetylenes
(1). This approach was initially investigated in a model study
by using stoichiometric amounts of Pd reagent and ligand.
R3OH (a, Scheme 1) could not simply be replaced by H2O
in order to generate carboxylic acids since the CO is readily
oxidized into formic acid in the presence of H2O, leading to
the palladium bleeding,5 which is indeed consistent with our
observation.
Table 1. Optimization of the Conditions
In the absence of the alcohol, our most efficient conditions
to generate benzo[b]furan-3-carboxylic esters (3) (a, Scheme
1) in the solution phase1c could not produce any acid 2a
(Table 1, entry 1), whereas the conditions we optimized for
synthesis of the 3 (a, Scheme 1) on solid phase1d did result
in the generation of the corresponding acid 2a in a low yield
(30%) with o-acetyloxyarylacetylene 3a (29%) as an unusual
byproduct (Table 1, entry 2). To increase the yield and
suppress this side reaction, various PdII species, counter-
anions, ligands, bases, and solvents were systematically
investigated (Table 1).
The solvent effect was first studied (Table 1, entries 2-5),
and CH3CN was found to provide the best result in
comparison with DMF, THF, or benzene.
Notably, bases other than the CsOAc (Table 1, entries 1,
6-9), such as organic bases selected for avoiding side
product 3a, blocked the reactions.
Among various PdII species investigated (Table 1, entries
10-13), Pd(CH3CN)2Cl2 was found to be the most reactive.
Various ligands were tested (Table 1, entries 14-19).
Obviously, electron-rich ligands are not favorable for in-
creasing the reactivity of the cationic PdII species. Finally,
2-PyPPh2, a highly efficient ligand for PdII-catalyzed Reppe
carbonylation,6 proved outstanding in our case, the yield of
2a dramatically being increased up to 70% and the byproduct
3a notably being suppressed to less than 10% (Table 1, entry
19).
To make the PdII species more cationic so as to further
increase its reactivity, AgOTs was added to extract Cl- from
the Pd(CH3CN)2Cl2 by forming AgCl and exchange Cl- with
a noncoordinating OTs-, which is a superior counteranion
for PdII.6,7 As a result, the side product 3a was completely
suppressed and the yield of 2a was increased to 80% (Table
1, entry 20). Similarly, AgBF4 was also tested, and a
comparable result was obtained (Table 1, entry 21). Further
reaction studies were performed by using a commercially
available cationic Pd(CH3CN)4(BF4)2 in the absence of any
silver salts, a satisfactory yield of 70% was achieved.
However, 10% of byproduct 3a was generated. Therefore,
based on our investigation, silver salt did play an important
role in suppressing the side reaction of acetylation.
By using the optimized conditions (Table 1, entry 20),
various o-hydroxyarylacetylenes with different substitution
patterns were transformed into their corresponding benzo-
[b]furan-3-carboxylic acids (2a-2g, Table 2), all in good
yields, which proved the significance of this novel method.
Many oxidants, such as O2, 1,4-benzoquinone, DDQ, Fe-
(OTs)3, CuCl2, Cu(OAc)2, CBr4, etc., were tried in order to
turn over the Pd0 back to PdII. Currently, the best result (40%
yield) was achieved by using Pd(CH3CN)4(BF4)2 (5%) and
1,4-benzoquinone (2 equiv).
(5) Fenton, D. M.; Steinwand, P. J. J. Org. Chem. 1974, 39, 701.
(6) (a) Drent, E.; Arnoldy, P.; Budzelaar, P. H. M. J. Orangomet. Chem.
1994, 475, 57. (b) Drent, E.; Arnoldy, P.; Budzelaar, P. H. M. J. Orangomet.
Chem. 1993, 455, 247.
(7) (a) Tsuji, J. Acc. Chem. Res. 1973, 6, 8. (b) Oi, S.; Nomura, M.;
Aiko, T.; Inoue, Y. J. Mol. Catal. A: Chem. 1997, 115, 289.
During our model study, two observations came to our
attention: one was the acetylation of substrate 1a (Table 1)
as a major side reaction, and the other was that bases other
than AcO- hardly promoted any formation of the product
2708
Org. Lett., Vol. 7, No. 13, 2005

Scheme 2. Plausible Mechanism
Table 2. Carboxylative Annulation on Diverse Substrates
MS studies. On the basis of this mechanism, if the PdII
species is not reactive enough to allow 1 to be consumed
rapidly, the 1 that remains will be acetylated by the anhydride
6 to give the unreactive acetate 3, which cannot be utilized
for further carboxylative annulation. Thus, our endeavor to
increase the reactivity of the cationic PdII fulfills the demands
acquired by this mechanism.
In summary, we have described a novel and mild method
for the rapid synthesis of benzo[b]furan-3-carboxylic acids
directly from the substituted o-alkynyl- phenols in good
yields by utilizing a PdII-mediated carboxylative annulation.
This method may find further application in the future for
rapidly constructing other useful heterocyclic carboxylic
acids.
2a (Table 1, entries 1, 6-8), which prompted us to propose
that AcO- does not simply serve as a base but is also
involved in the formation of both 2a and 3a. This led to our
newly proposed mechanism (Scheme 2). The PdII(CO)18OAc
could be generated from association of the original PdII
reagent with CO and H18OAc (generated from the deproto-
nation of the substrate 1 by Cs18OAc), and its further
association with 1 forms a triple-bond-activated PdII π-com-
plex 4. With the assistance of AcO-, the annulation of 4
could then give a PdIIσ-complex 5 and reductive elimination
could then ensue to generate the Pd0 and an acetic benzo-
[b]furan-3-carboxylic anhydride 6, which could serve as an
acetylating reagent for substrate 1 and lead to acetate 3 as a
byproduct. The hydrolysis of 6 during the workup would
afford the final product 2. This mechanism was supported
by an 18O-labeling study utilizing Cs18OAc followed by
quenching of the reaction with bulky t-BuOH in order to
achieve regioselective alcoholysis of the possible anhydride
intermediate 6 to obtain the 18O-labeled product 2, which
was isolated in high yield and confirmed by both NMR and
Acknowledgment. Inspiring discussions from Drs. Qiang
Zhu, Jie Wu and Youhong Hu, kind help from Ms. Yvonne
Pagano in preparing the Supporting Information, and valuable
supports from the VivoQuest, Inc. are gratefully acknowl-
edged.
Supporting Information Available: Experimental pro-
cedures and NMR and LC-MS spectra. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL050876G
Org. Lett., Vol. 7, No. 13, 2005
2709