been successful only with a limited class of electron-rich
heterocycles (i.e., indolizines2c and indoles2d). Herein, we
report the palladium-catalyzed direct alkynylation of benzene
derivatives with the aid of chelation assistance. Preliminary
mechanistic studies indicated that alkynylation would proceed
through a pathway that is distinctly different from that
reported for heterocycles.
Scheme 2
.
Pd-Catalyzed Direct Alkynylation of Aromatic C-H
Bonds with Bromoalkyne 2a
At the outset of our investigation, we observed that the
catalytic conditions developed for heterocycles (10 mol %
of PdCl2(PPh3)2, KOAc)2c,d were inapplicable to the direct
alkynylation of anilide 1 with bromoalkyne 2. We reasoned
that the electrophilicity of the postulated alkynylpalladium
species,2c generated by oxidative addition of 2, would be
insufficient to react with anilide 1, which is a considerably
poorer π-nucleophile than either indolines or indoles. We
postulated that the addition of silver salts would increase
the electrophilicity of the palladium species by sequestering
the bromide ligand, thus accelerating the palladation of 1,
as is frequently observed in other C-H bond functionaliza-
tion reactions.4 Indeed, the palladium-catalyzed reaction of
anilide 1 with bromoalkyne 2 furnished the expected ortho-
alkynylated product 3 in the presence of silver salts. After a
series of optimization studies,5 the yield was finally improved
to 70% (eq 1). Further investigation revealed that the choice
of the substituent on the alkynylating agent exerted a critical
impact on the reaction outcome. Replacing the triisopropyl-
silyl group in 2 with a tert-butyldimethylsilyl group led to a
significant reduction in yield (39%),6 while bromoalkynes
bearing other substituents, such as Ph, hexyl, and ester, did
not form the corresponding products, due in part to their
instability under these conditions. Although the scope
regarding the bromoalkyne component proved to be limited,
this does not deteriorate the utility of the reaction severely,
since the triisopropylsilyl group at the alkyne terminus can
readily be deprotected and elaborated (vide infra).
a Reaction conditions: anilide (0.5 mmol), bromoalkyne 2 (0.75 mmol),
Pd(OAc)2 (0.05 mmol), AgOTf (0.5 mmol), and K2CO3 (0.50 mmol) in
toluene (1.0 mL) at 70 °C, 15 h. Isolated yields based on anilides are shown.
b Run at 50 °C.
larenes in good yields (4, 5, 8, 9, 12, and 13 in Scheme 2) and
formed no dialkynylated products.7 On the other hand, dimin-
ished but still synthetically acceptable yields were obtained with
anilides bearing electron-withdrawing groups (6, 7, 10, and 11).
These results agreed with the general reactivity trend observed
in C-H functionalization reactions involving electrophilic
palladation.8 The negative effect of electron-withdrawing groups
was offset when such substrates were accompanied by an
electron-donating group (14 and 15). These examples highlight
the synthetic advantage of the direct alkynylation method
compared with the Sonogashira-Hagihara reaction, in which
the preparation of complicated aryl halides is required. Regard-
ing the directing groups, cyclic amides (16) and acetanilide with
an unprotected N-H group also efficiently underwent alkyny-
lation,9 although under the catalytic conditions in the latter case,
the primary product 17 was susceptible to hydrolytic cleavage
of a C-Si bond to afford 18.10
Under the catalytic conditions established above, we next
explored the scope of anilides. Electron-rich substrates bearing
alkyl or methoxy groups furnished the corresponding alkyny-
To probe the nature of the C-H bond cleavage, we next
investigated the kinetic isotope effect of this direct alkyny-
lation reaction (Scheme 3). Both intra- and intermolecular
(4) Lebrasseur, N.; Larrosa, I. J. Am. Chem. Soc. 2008, 130, 2926, and
references therein.
(5) Notes: (a) Other palladium sources tested: PdCl2 (trace),
Pd2(dba)3·CHCl3 (67%), Pd(OCOCF3)2 (58%), Pd(OAc)2/2PPh3 (53%) . (b)
Other silver salts tested: AgBF4 (57%), AgSbF6 (13%), AgOAc (0%),
Ag2CO3 (0%). (c) Other bases tested: KOAc (trace), K3PO4 (38%), Cs2CO3
(trace), Et3N (39%). (d) The corresponding chloro- and iodoalkynes afforded
3 in 2% and 48% yield, respectively.
(7) Difunctionalization is often a problem in chelation-assisted direct
functionalization of C-H. For example, see: Giri, R.; Maugel, N.; Li, J.-J.;
Wang, D.-H.; Breazzano, S. P.; Saunders, L. B.; Yu, J.-Q. J. Am. Chem.
Soc. 2007, 129, 3510.
(8) Selected examples: (a) Yang, S.; Li, B.; Wan, X.; Shi, Z. J. Am.
Chem. Soc. 2007, 129, 6066. (b) Shi, Z.-J.; Li, B.-J.; Wan, X.; Cheng, J.;
Fang, Z.; Cao, B.; Qin, C.; Wang, Y. Angew. Chem., Int. Ed. 2007, 46,
5554. (c) Giri, R.; Yu, J.-Q. J. Am. Chem. Soc. 2008, 130, 14082.
(9) N,N-Dimethylaminomethyl, 2-pyridyl, N,N-dimethylaminocarbonyl,
and acetoxy groups did not serve as directing groups.
(6) Superior performance of Si(i-Pr)3-protected alkynes was also reported
in other catalysis: (a) Tsukada, N.; Ninomiya, S.; Aoyama, Y.; Inoue, Y.
Org. Lett. 2007, 9, 2919. (b) Nishimura, T.; Guo, X.-X.; Uchiyama, N.;
Katoh, T.; Hayashi, T. J. Am. Chem. Soc. 2008, 130, 1576. (c) Shirakura,
M.; Suginome, M. J. Am. Chem. Soc. 2008, 130, 5410. (d) Ogata, K.;
Murayama, H.; Sugasawa, J.; Suzuki, N.; Fukuzawa, S.-i. J. Am. Chem.
Soc. 2009, 131, 3176.
(10) Compound 17 was converted into 18 in the presence of TfOH. See
the Supporting Information for details.
Org. Lett., Vol. 11, No. 15, 2009
3251