Synthesis of aryl ketones or ketimines by palladium-catalyzed arene C-H addition to nitriles

Article abstract of DOI:10.1021/jo060220g

The unprecedented palladium-catalyzed C-H addition of arenes to nitriles provides moderate to excellent yields of aryl ketones or the corresponding hindered imines. The addition of a small amount of DMSO increases the yields dramatically. Both intermolecular and intramolecular reactions are successful, although the intramolecular reactions tend to be more sluggish. This novel chemistry is believed to involve palladium-catalyzed C-H activation of the arene by electrophilic aromatic substitution, followed by the unusual carbopalladation of a nitrile. Similar reactions have been successfully developed employing arylboronic acids and nitriles. A concise route to xanthones starting from cheap starting materials has been developed employing this synthetic protocol.

Full text of DOI:10.1021/jo060220g

Synthesis of Aryl Ketones or Ketimines by Palladium-Catalyzed
Arene C-H Addition to Nitriles
Chengxiang Zhou and Richard C. Larock*
Department of Chemistry, Iowa State UniVersity, Ames, Iowa 50011
larock@iastate.edu
ReceiVed February 1, 2006
The unprecedented palladium-catalyzed C-H addition of arenes to nitriles provides moderate to excellent
yields of aryl ketones or the corresponding hindered imines. The addition of a small amount of DMSO
increases the yields dramatically. Both intermolecular and intramolecular reactions are successful, although
the intramolecular reactions tend to be more sluggish. This novel chemistry is believed to involve
palladium-catalyzed C-H activation of the arene by electrophilic aromatic substitution, followed by the
unusual carbopalladation of a nitrile. Similar reactions have been successfully developed employing
arylboronic acids and nitriles. A concise route to xanthones starting from cheap starting materials has
been developed employing this synthetic protocol.
Introduction
acetonitrile is a commonly employed solvent in organopalladium
chemistry. Up to now, only a few examples of reactions
involving the intramolecular carbopalladation of nitriles have
been reported.6,7 Recently, we communicated the first example
of Pd-catalyzed simple C-H addition to nitriles, which provides
a useful new synthesis of aromatic ketones or ketimines (eq
1).8 Herein, we wish to provide a full account of the scope and
Developing reliable methods for catalytic C-H functional-
ization has been a long-term goal in synthetic organic chemistry.
For example, the catalytic addition of C-H bonds to unsaturated
substrates has been one of the most attractive research subjects
for synthetic chemists for years.1 The late transition metals, such
as Pd, have been shown to catalyze C-H addition to simple
alkenes and alkynes and provide various valuable synthetic
intermediates in an atom-economical manner.2 To our knowl-
edge, however, there are no examples of Pd-catalyzed C-H
additions to polar multiple bonds, such as a nitrile.3,4 In fact,
the nitrile functionality remains untouched during most Pd-
catalyzed organic reactions.5 Indeed, the nitrile complexes,
PdCl2(RCN)2 (R ) Me, Ph), are widely used Pd catalysts and
limitations of this chemistry. We have also developed a useful
ketone synthesis by the reaction of arylboronic acids and nitriles
(5) (a) Tsuji, J. Palladium Reagents and Catalysis: New PerspectiVes
for the 21st Century; John Wiley & Sons: New York, 2004. (b) Handbook
of Organopalladium Chemistry for Organic Synthesis; Negishi, E., Ed.; John
Wiley & Sons: New York, 2002.
(1) For selective reviews, see: (a) Shilov, A. E.; Shul’pin, G. B. Chem.
ReV. 1997, 97, 2879. (b) Labinger, J. A.; Bercaw, J. E. Nature 2002, 417,
507. (c) Kakiuchi, F.; Chatani, N. AdV. Synth. Catal. 2003, 345, 1077.
(2) For selective examples, see: (a) Tsukada, N.; Mitsuboshi, T.;
Setoguchi, H.; Inoue, Y. J. Am. Chem. Soc. 2003, 125, 12102. (b) Boele,
M. D. K.; van Strijdonck, G. P. F.; de Vries, A. H. M.; Kamer, P. C. J.; de
Vries, J. G.; van Leeuwen, P. W. N. M. J. Am. Chem. Soc. 2002, 124,
1586. (c) Ritleng, V.; Sirlin, C.; Preffer, M. Chem. ReV. 2002, 102, 1731
and references therein.
(3) For a review covering metal-activated organonitriles, see: Kukushkin,
V. Y.; Pombeiro, A. J. L. Chem. ReV. 2002, 102, 1771.
(4) For recent rhodium-catalyzed nucleophilic additions to nitriles, see:
(a) Miura, T.; Murakami, M. Org. Lett. 2005, 7, 3339. (b) Miura, T.;
Nakazawa, H.; Murakami, M. Chem. Commun. 2005, 2855. (c) Ueura, K.;
Satoh, T.; Miura, M. Org. Lett. 2005, 7, 2229.
(6) (a) Larock, R. C.; Tian, Q.; Pletnev, A. A. J. Am. Chem. Soc. 1999,
121, 3238. (b) Pletnev, A. A.; Larock, R. C. Tetrahedron Lett. 2002, 43,
2133. (c) Pletnev, A. A.; Tian, Q.; Larock, R. C. J. Org. Chem. 2002, 67,
9276. (d) Pletnev, A. A.; Larock, R. C. J. Org. Chem. 2002, 67, 9428. (e)
Tian, Q.; Pletnev, A. A.; Larock, R. C. J. Org. Chem. 2003, 68, 339. (f)
Yang, C. C.; Tai, H. M.; Sun, P. J. J. Chem. Soc., Perkin Trans. 1 1997,
2843. (g) Yang, C. C.; Sun, P. J.; Fang, J. M. J. Chem. Soc., Chem. Commun.
1994, 2629. (h) Luo, F. H.; Chu, C. I.; Cheng, C. H. Organometallics 1998,
17, 1025. (i) Zhao, L.; Lu, X. Angew. Chem., Int. Ed. 2002, 41, 4343.
(7) For a Pd(II)-assisted Ritter reaction, see: Hegedus, L. S.; Mulhern,
T. A.; Asada, H. J. Am. Chem. Soc. 1986, 108, 6224.
(8) Zhou, C.; Larock, R. C. J. Am. Chem. Soc. 2004, 126, 2302.
10.1021/jo060220g CCC: $33.50 © 2006 American Chemical Society
Published on Web 04/04/2006
J. Org. Chem. 2006, 71, 3551-3558
3551

Zhou and Larock
TABLE 1. Optimization Studies (Eq 2)a
with the hope of stabilizing the possible Pd(II) intermediate
while maintaining the highly cationic palladium intermediate,
had little effect on the reaction (entries 3 and 4). However, the
addition of DMSO afforded a large increase in the yield (entries
5-10). A 76% yield of the desired product can be obtained
when 2 equiv of DMSO per nitrile is employed (entry 7). More
or less, DMSO gave lower yields. It is noteworthy that the
reaction can even be improved dramatically by adding only a
catalytic amount of DMSO (compare entries 1 and 9). A slightly
lower yield is obtained when only 5% Pd(OAc)2 is employed
(compare entries 7 and 10). The use of TFA as solvent is
apparently crucial for the success of this chemistry because none
of the desired product was obtained when acetic acid was
employed as the solvent (entry 11). Other Pd catalysts, such as
PdCl2(PPh3)2, Pd(PPh3)4, and Pd/C, are ineffective in this
chemistry (entries 12-14). The Pd(II) catalyst is uniquely
effective because other late transition metal salts or common
Lewis acids all failed to give any of the desired product (entries
15-22). No reaction occurs in the absence of the catalyst (entry
23). Thus, the optimal procedure described in entry 7 of Table
1 has been employed to study the scope of this chemistry.
(b) Reaction of Various Nitriles and Arenes. Using the
optimized reaction conditions, a number of arenes and nitriles
have been successfully employed in this ketone synthesis (Table
2). The reaction of benzonitrile and benzene afforded a much
lower yield of the desired product (compare entries 1 and 2),
possibly because of the fact that benzene is less electron rich
than toluene and benzene is rather volatile at the reaction
temperature. The reactions of benzonitrile and relative electron-
rich p-xylene (entry 3), anisole (entry 4), 1,4-dimethoxybenzene
(entry 5), and 1,3,5-trimethoxybenzene (entry 6) have all
afforded the corresponding ketones in good yields. It is
noteworthy that the ortho and para isomers 4a and 4b were
obtained in a ratio of 52:48 when anisole was employed (entry
4). The arenes 4-tert-butylphenol and 4-methylphenol cleanly
gave the acylation products ortho to the hydroxyl group in good
yields (entries 7 and 8). The arene 4-bromophenol also afforded
the acylation product ortho to the hydroxyl group, although in
a much lower yield (entry 9). None of the desired product was
obtained when 4-iodophenol was employed, possibly because
this substrate is unstable in TFA under the reaction conditions
or because oxidative addition of the C-I bond to the Pd
intermediate disrupts our desired chemistry (entry 10). The arene
4-methoxyphenol also failed to give any of the desired product,
possibly because of oligomerization (entry 11). Mesitylene is
more reactive, affording good yields at only 65 or 75 °C (entries
12 and 13). Interestingly, biaryl ketimines, instead of ketones,
are obtained as the only products in these reactions. Other
sterically hindered arenes, such as bromomesitylene and 1,2,4,5-
tetramethylbenzene have also afforded the corresponding
ketimines as products (entries 14 and 15). The steric hindrance
around the imine group apparently hinders the hydrolysis
process.10 The relatively electron-poor p-BrC6H4CN reacts faster
and leads to a higher yield of ketimine than the more electron-
rich p-MeOC6H4CN (compare entries 16 and 17). Various
halogen-containing benzonitriles are excellent substrates for this
chemistry, providing the corresponding ketimines in good to
excellent yields (entries 17-22). For example, the reaction of
1,3,5-trimethoxybenzene and 2-fluorobenzonitrile afforded the
corresponding ketimine in 89% yield (entry 21), compared with
entry
catalyst
additiveb
% yieldc
1
2e
3
4
5
6
7
8
9
10% Pd(OAc)2
-
-
10d
0
20 (16)
15
10% Pd(OAc)2
10% Pd(OAc)2
10% Pd(OAc)2
10% Pd(OAc)2
10% Pd(OAc)2
10% Pd(OAc)2
10% Pd(OAc)2
10% Pd(OAc)2
5% Pd(OAc)2
10% Pd(OAc)2
10% PdCl2(PPh3)2
10% Pd(PPh3)4
10% Pd/C
10 equiv of THF
10 equiv of PhNO2
10 equiv of DMSO 45
5 equiv of DMSO
2 equiv of DMSO
1 equiv of DMSO
0.4 equiv of DMSO 55
2 equiv of DMSO
2 equiv of DMSO
56
76 (68)d
66
10f
11g
12
13
14
15
16
17
18
19
20
21
22
23
61
0
10 equiv of DMSO trace
10 equiv of DMSO
10 equiv of DMSO
2 equiv of DMSO
0
0
0
0
0
0
0
0
0
0
0
10% PtCl2 + 20% AgO2CCF3
10% RhCl3 + 30% AgO2CCF3 2 equiv of DMSO
20% Hg(O2CCF3)2
20% AlCl3
20% SnCl4
20% TiCl4
20% Mg(OAc)2
20% Zn(OAc)2
-
2 equiv of DMSO
2 equiv of DMSO
2 equiv of DMSO
2 equiv of DMSO
2 equiv of DMSO
2 equiv of DMSO
2 equiv of DMSO
a Unless otherwise indicated, all reactions were run by employing 2.0
mmol of toluene and 1.0 mmol of benzonitrile in the presence of the catalyst
indicated in 2.5 mL of TFA at 90 °C for 24 h, followed by hydrolysis.
b The equivalents of additives are based on benzonitrile. c GC yields; yields
of products obtained by column chromatography are reported in parentheses.
d Ortho, meta, and para isomers were obtained in a ratio of 51:16:33. e The
reaction was run at room temperature. f The reaction was run for 48 h. g The
reaction was run employing AcOH as the solvent, instead of TFA.
under similar reaction conditions. This chemistry has also
provided a concise route to xanthones starting from cheap
starting materials.
Results and Discussion
(a) Optimization of the Reaction Conditions. Recently,
Fujiwara reported highly efficient Pd-catalyzed C-H additions
to various alkynes and alkenes in TFA.9 The highly cationic
Pd(II) species, generated in TFA, is believed to be the key to
the efficiency of this chemistry. We envisioned that such a
highly cationic arylpalladium species, when generated in situ
by the electrophilic palladation of arenes, should strongly
coordinate with a nitrile and perhaps facilitate the “abnormal”
carbopalladation of a nitrile. To this end, we initially studied
the Pd-catalyzed reaction of toluene and benzonitrile in TFA
(eq 2).
To our delight, a 10% yield of the desired products (o/m/p
) 51:16:33) was obtained by employing 2 mmol of toluene, 1
mmol of benzonitrile, and 0.1 mmol of Pd(OAc)2 in 2.5 mL of
TFA at 90 °C for 24 h (entry 1, Table 1). No reaction occurs at
room temperature (entry 2). Adding either THF or nitrobenzene,
(9) (a) Jia, C.; Piao, D.; Oyamada, J.; Lu, W.; Kitamura, T.; Fujiwara,
Y. Science 2000, 287, 1992. (b) Jia, C.; Lu, W.; Oyamada, J.; Kitamura,
T.; Matsuda, K.; Irie, M.; Fujiwara, Y. J. Am. Chem. Soc. 2000, 122, 7252.
(c) Jia, C.; Kitamura, T.; Fujiwara, Y. Acc. Chem. Res. 2001, 34, 633 and
references therein.
(10) The sterically hindered biaryl ketimines cannot be hydrolyzed even
after heating at 100 °C for 24 h in aqueous HCl or K2CO3 solutions.
3552 J. Org. Chem., Vol. 71, No. 9, 2006

Synthesis of Aryl Ketones or Ketimines
TABLE 2. Pd-Catalyzed Reaction of Nitriles and Arenes (Eq 1)a
a Unless indicated otherwise, all reactions were run by employing 2.0 mmol of arene, 1.0 mmol of nitrile, 0.10 mmol of Pd(OAc)2, and 0.10 mL of
DMSO in 2.5 mL of TFA for 24 h. The reactions were then worked up under acidic or basic conditions (see the Experimental Section for details). b Isolated
yields. c Ortho, meta, and para isomers were obtained in a ratio of 51:16:33. d Ortho and para isomers were obtained in a ratio of 52:48. e The reaction was
run for 48 h. f The reaction employed 0.20 mmol of nitrile, 0.03 mmol of Pd(OAc)2, 0.20 mL of DMSO, and 5.0 mL of TFA for 36 h.
J. Org. Chem, Vol. 71, No. 9, 2006 3553

Zhou and Larock
a 66% yield of the ketone product obtained when 1,3,5-
trimethoxybenzene and benzonitrile were employed (entry 6).
A ketone was obtained when 1,4-dimethoxybenzene was
employed with this latter nitrile (entry 22). It is important to
note that acetonitrile also worked well in this chemistry (entry
23). Unfortunately, an indole failed to give any of the desired
product (entry 24), possibly because the substrate is not stable
under our reaction conditions. Interestingly, intramolecular
variations of this reaction are more difficult than the inter-
molecular reactions. For example, only a 16% yield of the
xanthone product 24 was obtained when σ-phenoxybenzonitrile
was employed (entry 25). Fortunately, a seven-membered ring
product 25 was formed more readily, although a higher
temperature was required (entry 26). Simple phenyl-containing
alkanenitriles failed to generate the expected intramolecular
cyclization products (entries 27-29). However, introducing two
methyl groups on the carbon adjacent to the nitrile facilitated
cyclization to the corresponding six- and seven-membered ring
products 30 and 31 in moderate yields (entries 31 and 32),
possibly because of the gem-disubstituent effect.11 Noteworthy
is the fact that the formation of five- and six-membered ring
ketones is more difficult than formation of a seven-membered
ring ketone (compare entries 30, 31, and 32). This unusual trend
suggests that this reaction is most likely not a Lewis-acid-
catalyzed Houben-Hoesch reaction12 (see the Reaction Mech-
anism section). In fact, none of the aboVe reactions in Table 2
proVide any of the desired products without the Pd catalyst.
obtained previously by our ketone synthesis (see entry 22 in
Table 2), followed by cyclization by K2CO3 (eq 4).
(d) Reaction of Nitriles and Arylboronic Acids. The success
of the Pd-catalyzed reaction of arenes and nitriles suggests that
a similar reaction should occur with nitriles if an arylpalladium
species is generated by transmetalation.18 Indeed, we have found
that arylboronic acids react under our standard reaction condi-
tions with nitriles to provide the corresponding ketimine or
ketone products (eq 5, Table 4).
Good to excellent yields have been obtained when electron-
rich arylboronic acids react with benzonitrile (entries 1-4).
Small amounts of ortho and meta isomers were also obtained
when p-tolylboronic acid was employed (footnote c in entry
2). Keep in mind that a 50:12:38 o/m/p ratio of ketone products
was obtained in 62% yield when toluene was allowed to react
with 5 equiv of PhCN under these same reaction conditions.
The minor amounts of o and m isomers may be arising by an
alternate protonolysis process, which generates toluene, which
subsequently undergoes direct reaction with benzonitrile. Thus,
this arylboronic acid approach to aromatic ketones solves some
of the regiochemical problems observed when simple arenes
are employed. Similarly, an 8:92 o/p ratio of ketones was
obtained when p-methoxyphenylboronic acid was employed
(footnote d in entry 3), whereas a 52:48 ratio was observed when
anisole was employed directly in this process. The reaction of
mesitylboronic acid and p-bromobenzonitrile afforded the
ketimine product in a good yield (entry 5), as opposed to the
Suzuki coupling product one would normally expect if this
reaction was run in the presence of a base. A 38% yield of the
desired ketone product was obtained when electron-poor p-
nitrophenylboronic acid was employed (entry 6). Unfortunately,
none of the desired product was obtained when 2-thienylboronic
acid was used, possibly because of the instability of this boronic
acid in TFA (entry 7).
(c) Two-Step Approach to Xanthones. Xanthones are
abundant in numerous natural products and possess many
important biological activities.13 As an application of our
chemistry, a simple approach to xanthones has been developed,
which involves ketone formation using simple phenols and
2-fluorobenzonitrile as starting materials, followed by intra-
molecular cyclization under mild conditions (eq 3).
Various xanthones have been obtained from cheap, readily
available starting materials in moderate yields (Table 3, entries
1-4). It is noteworthy that a chloride-containing xanthone was
obtained (entry 4) because this substrate should be subject to
facile functionalization via various Pd-catalyzed reactions.14
4-Phenylphenol failed to give the ketone product, possibly
because of solubility problems with this phenol in TFA (entry
5). 4-Methoxyphenol also failed to give the desired ketone
product (entry 6). However, xanthone 37 can be obtained in
68% overall yield via selective demethylation of the ketone 21
Reaction Mechanism. A plausible mechanism for this ketone
synthesis is illustrated in Scheme 1. It involves the following
key steps: (1) electrophilic metalation of the arene by the
Pd(II) catalyst A,15 which generates arylpalladium species B;9,16
(2) coordination of the nitrile to the Pd; (3) carbopalladation of
the nitrile to form the imine-Pd(II) complex C;17 (4) protonation
of C by TFA, which affords the ketimine product D and
(11) For a review, see: Jung, M. E.; Piizzi, G. Chem. ReV. 2005, 105,
1735.
(12) For the Houben-Hoesch reaction, see: (a) Spoerri, P. E.; DuBois,
A. S. Org. React. 1949, 5, 387. (b) Sato, Y.; Yato, M.; Ohwada, T.; Saito,
S.; Shudo, K. J. Am. Chem. Soc. 1995, 117, 3037.
(15) The complex (DMSO)2Pd(O2CCF3)2 has been well characterized;
see: Bancroft, D. P.; Cotton, F. A.; Verbruggen, M. Acta Crystallogr. Sect.
C 1989, 45, 1289.
(13) For selective examples, see: (a) Schwaebe, M. K.; Moran, T. J.;
Whitten, J. P. Tetrahedron Lett. 2005, 46, 827. (b) Mulholland, D. A.;
Koorbanally, C.; Crouch, N. R.; Sandor, P. J. Nat. Prod. 2004, 67, 1726.
(c) Kenji, M.; Yukihiro, A.; Hong, Y.; Kenji, O.; Tetsuro, I.; Toshiyuki,
T.; Emi, K.; Munekazu, I.; Yoshinori, N. Bioorg. Med. Chem. 2004, 12,
5799. (d) Pedro, M.; Cerqueira, F.; Sousa, M. E.; Nascimento, M. S. J.;
Pinto, M. Bioorg. Med. Chem. 2002, 10, 3725 and references therein.
(14) For a review of the Pd-catalyzed functionalization of aryl chlorides,
see: Littke, A. F.; Fu, G. C. Angew Chem., Int. Ed. 2002, 41, 4176.
(16) For the formation of arylpalladium complexes by electrophilic
palladation of arenes in TFA, see also: (a) Lu, W.; Yamaoka, Y.; Taniguchi,
Y.; Kitamura, T.; Fujiwara, Y. J. Organomet. Chem. 1999, 586, 290. (b)
Fuchita, Y.; Hiraki, K.; Kamogawa, Y.; Suenaga, M.; Toggoh, K.; Fujiwara,
Y. Bull. Chem. Soc. Jpn. 1989, 62, 1081. (c) Gretz, E.; Oliver, T. F.; Sen,
A. J. Am. Chem. Soc. 1987, 109, 8109. (d) Clark, F. R. S.; Norman, R. O.
C.; Thommas, C. B.; Willson, J. S. J. Chem. Soc., Perkin Trans. 1 1974,
1289.
3554 J. Org. Chem., Vol. 71, No. 9, 2006

Synthesis of Aryl Ketones or Ketimines
TABLE 3. Synthesis of Xanthones (Eq 3)
a Unless indicated otherwise, all reactions were run by employing 2.5 mmol of phenol, 1.0 mmol of 2-fluorobenzonitrile, 0.10 mmol of Pd(OAc)2, and
0.10 mL of DMSO in 2.5 mL of TFA for 24 h. The reactions were then worked up under acidic or basic conditions, and the products were purified by
column chromatography. b The reactions were run by employing 0.25 mmol of ketone and 0.50 mmol of K2CO3 in acetone at 50 °C for 2 h. c Isolated yields.
regenerates the Pd(II) catalyst. On the other hand, the aryl-
palladium species B can undergo further electrophilic metalation
with another molecule of arene and subsequent reductive
elimination can produce the biaryl side product sometimes
observed in our reactions.
Alternatively, a Pd(II)-catalyzed Houben-Hoesch-like12 mech-
anism is also possible as shown in Scheme 2. The key steps
here would be: (1) coordination of the nitrile to Pd(II); (2)
nucleophilic attack of the arene on the activated nitrile to
generate a cationic intermediate; (3) deprotonation to generate
imine-Pd(II) complex C; and (4) protonation of C by TFA,
which affords the ketimine product D and regenerates the Pd-
(II) catalyst.
tion of the preference for o- and p-isomers over the m-isomer
in the products from toluene or anisole is consistent with
electrophilic palladation of the arene9 (entries 1 and 4, Table
2). Facile carbopalladation of the nitrile should arise from the
highly cationic arylpalladium complex B generated under our
reaction conditions.17 The carbopalladation step might involve
nucleophilic attack of the arylpalladium intermediate B on the
nitrile, which is consistent with the observation that faster
reactions and higher yields are obtained when using the more
electron-deficient p-BrC6H4CN over the more electron-rich
p-MeOC6H4CN (entries 16 and 17, Table 2).
(B) The Pd(II) catalyst plays a unique role in the success of
this chemistry. The combination of a very small amount of
Pd(II) and TFA should not be enough to induce the Houben-
Hoesch reaction given the fact that the pKa of RCN‚H+ is around
-10 and a stronger acidic media should be necessary to induce
a Houben-Hoesch reaction.12b Many other strong Lewis acids,
including AlCl3 and SnCl4, do not catalyze this transformation
as indicated in Table 1. These results strongly suggest that the
Houben-Hoesch mechanism in Scheme 2 is not operable in
our reaction.
We strongly favor the mechanism in Scheme 1 over the latter
mechanism because of the following observations.
(A) The key mechanistic steps proposed in Scheme 1 all have
precedent. The electrophilic palladation of arenes under similar
conditions is well documented in the literature.9,16 Our observa-
(17) For a recent example involving the insertion of a nitrile into a Pd-C
bond, see: Vicente, J.; Abad, J. A.; Lopez-Saez, M.; Jones, P. G. Angew.
Chem., Int. Ed. 2005, 44, 6001.
J. Org. Chem, Vol. 71, No. 9, 2006 3555

Zhou and Larock
TABLE 4. Pd-Catalyzed Reaction of Nitriles and Arylboronic Acids (Eq 5)a
a Unless indicated otherwise, all reactions were run by employing 1.0 mmol of arylboronic acid, 5.0 mmol of nitrile, 0.10 mmol of Pd(OAc)2, and 0.10
mL of DMSO in 2.5 mL of TFA at 90 °C for 24 h. The reactions were then worked up under acidic or basic conditions. b Isolated yields. c Ortho, meta, and
para isomers were obtained in a ratio of 9:3:88. d Ortho and para isomers were obtained in a ratio of 8:92.
(C) If a Houben-Hoesch mechanism operates, one would
expect the intramolecular reactions to be more facile than the
intermolecular reactions because of easier access to the arene.
However, the intramolecular variation of the reaction is much
more sluggish than the intermolecular variation as described
earlier. A unique trend for the intramolecular variation (seven-
over six- and five-membered ring formations) is observed, which
is inconsistent with a Houben-Hoesch mechanism. However,
the more sluggish formation of five- and six-membered ring
ketones over seven-membered ring ketones by intramolecular
variations of our palladium process may be explained by the
mechanism shown in Scheme 1 because it is more difficult for
the anticipated arylpalladium intermediate to form a stable cyclic
intermediate in which the Pd is also coordinated with the pair
of electrons on the nitrile nitrogen in these smaller ring systems.
(D) If a Houben-Hoesch mechanism operates, one would
expect a sterically hindered arene to encounter difficulties
approaching the nitrile, thus leading to more sluggish reactions.
However, this contradicts the fact that the sterically hindered
mesitylene is apparently much more reactive than toluene in
our Pd reactions.
(E) The mechanism in Scheme 1 suggests that the aryl-
palladium species B, if generated by some other means, should
also react with nitriles under similar reaction conditions. The
success of the reactions between arylboronic acids and nitriles
under similar reaction conditions strongly supports this C-H
activation mechanism. We believe that transmetalation between
the arylboronic acid and the Pd(II) catalyst should generate such
arylpalladium species under our reaction conditions (Scheme
3).18
The presence of DMSO greatly increases the yield of the
reaction. It is not clear what exactly the role of the DMSO is in
(18) For recent examples involving the transmetalation of arylboronic
acids by Pd(II), see: (a) Nishikata, T.; Yamamoto, Y.; Miyaura, N. Angew.
Chem., Int. Ed. 2003, 42, 2768. (b) Oh, C.; Jung, H.; Kim, K.; Kim, N.
Angew. Chem., Int. Ed. 2003, 42, 805. (c) Zhou, C.; Larock, R. C. Org.
Lett. 2005, 7, 259.
3556 J. Org. Chem., Vol. 71, No. 9, 2006

Synthesis of Aryl Ketones or Ketimines
SCHEME 1. Proposed Mechanism
of this reactive species and suppressing its further reaction with
another molecule of arene. On the other hand, DMSO might
also facilitate the reoxidation of Pd(0) to Pd(II) by air22 should
the Pd(II) ever be reduced to Pd(0) by the arenes.
Conclusions
The unprecedented palladium-catalyzed C-H addition of
arenes to nitriles provides moderate to excellent yields of aryl
ketones or the corresponding hindered ketimines. The addition
of a small amount of DMSO increases the yields dramatically.
Both intermolecular and intramolecular examples of this process
are successful. This novel chemistry is believed to involve
palladium-catalyzed C-H activation of the simple arene,
followed by an unusual carbopalladation of the nitrile. Similar
reactions have been successfully developed employing aryl-
boronic acids and nitriles. A concise route to xanthones starting
from cheap starting materials has been developed employing
this synthetic protocol.
Experimental Section
Intermolecular Reaction of Nitriles and Arenes (Table 2,
Entries 1-24). The arene (2.0 mmol), the nitrile (1.0 mmol),
Pd(OAc)2 (0.10 mmol), DMSO (0.10 mL), and TFA (2.5 mL) were
placed in a 6 dram vial. The vial was sealed, and the contents were
stirred and heated at the indicated temperature for 24 h. The ketone
products were obtained using the following acid workup procedure,
and the ketimine products were obtained by using the following
basic workup procedure.
SCHEME 2. Proposed Mechanism
(a) Acidic workup. Water (15 mL) was added to the vial, and
the resulting mixture was heated at 70 °C for 2 h. The mixture was
then cooled and extracted with diethyl ether three times. The
combined organic layers were dried over anhydrous MgSO4, and
the solvent was evaporated under reduced pressure. The crude
product was purified by chromatography on a silica gel column.
(b) Basic Workup. The reaction mixture was cooled to room
temperature, and then K2CO3 was added to the vial until no CO2
bubbles were generated (Caution! A large amount of CO2 is
generated because of the reaction of TFA and K2CO3). The resulting
mixture was extracted with diethyl ether three times. The combined
organic layers were dried over anhydrous MgSO4, and the solvent
was evaporated under reduced pressure. The crude product was
purified by chromatography on a silica gel column.
Intramolecular Reaction of Arylalkanenitriles (Table 2,
Entries 25-32). The nitrile (0.2 mmol), Pd(OAc)2 (0.03 mmol),
DMSO (0.2 mL), and TFA (5.0 mL) were placed in a 6 dram vial.
The vial was sealed, and the contents were stirred and heated at
the indicated temperature for 24 h. The resulting mixture was then
worked up according to the above acidic workup procedure.
General Procedure for the Intramolecular Cyclization Lead-
ing to Xanthones (Step 2 in Eq 3). The (2-fluorophenyl)(2-
hydroxyphenyl)methanone (0.25 mmol), K2CO3 (0.50 mmol), and
acetone (15 mL) were placed in a round-bottom flask. The flask
was sealed, and the contents were stirred at 50 °C for 2 h. The
suspension was then filtered through a pad of Celite and washed
with 50 mL of ethyl ether. The filtrate was concentrated to afford
the desired xanthones.
SCHEME 3. Transmetalation
the reaction.19 DMSO is known to be a unique ligand in many
useful Pd(II) transformations.20 We propose that the DMSO-
coordinated complex A15 effectively undergoes electrophilic
C-H activation of the arene to generate a DMSO-stabilized
arylpalladium intermediate B.21 The coordination of DMSO in
B might stabilize this Pd(II) intermediate, increasing the lifetime
(19) Our efforts to isolate the DMSO-coordinated palladium intermediates
have been unsuccessful.
Reaction of Nitriles and Arylboronic Acids (Table 4). The
nitrile (5.0 mmol), the arylboronic acid (1.0 mmol), Pd(OAc)2 (0.10
(20) For other examples of the importance of DMSO as a solvent or
ligand in Pd-catalyzed processes, see: (a) Larock, R. C.; Hightower, T. R.
J. Org. Chem. 1993, 58, 5298. (b) van Benthem, R. A. T. M.; Hiemstra,
H.; Michels, J. J.; Speckamp, W. N. J. Chem. Soc., Chem. Commun. 1994,
357. (c) Myers, A. G.; Tanaka, D.; Mannion, M. R. J. Am. Chem. Soc.
2002, 124, 11250. (d) Chen, M. S.; White, M. C. J. Am. Chem. Soc. 2004,
126, 1346. (e) Fraunhoffer, K. J.; Bachovchin, D. A.; White, M. C. Org.
Lett. 2005, 7, 223. (f) Chen, M. S.; Prabagaran, N.; Labenz, N. A.; White,
M. C. J. Am. Chem. Soc. 2005, 127, 6970.
(21) DMSO-coordinated arylpalladium complexes have been character-
ized. See: Tanaka, D.; Romeril, S. P.; Myers, A. G. J. Am. Chem. Soc.
2005, 127, 10323.
(22) (a) Stahl, S. S. Angew. Chem., Int. Ed. 2004, 43, 3400 and references
therein. (b) Steinhoff, B. A.; Fix, S. R.; Stahl, S. S. J. Am. Chem. Soc.
2002, 124, 766. (c) Peterson, K. P.; Larock, R. C. J. Org. Chem. 1998, 63,
3185.
J. Org. Chem, Vol. 71, No. 9, 2006 3557

Zhou and Larock
mmol), DMSO (0.10 mL), and TFA (2.5 mL) were placed in a 6
dram vial. The vial was sealed, and the contents were stirred and
heated at 90 °C for 24 h. The resulting mixture was then worked
up according to the above acidic or basic workup procedures.
cals Company Ltd. for providing the palladium compounds and
Frontier Scientific Company for the arylboronic acids.
Supporting Information Available: Experimental details and
1
product characterization data and H and 13C NMR spectra for all
new compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. Partial financial support from the Na-
tional Science Foundation is gratefully acknowledged. We also
acknowledge Johnson Matthey, Inc. and Kawaken Fine Chemi-
JO060220G
3558 J. Org. Chem., Vol. 71, No. 9, 2006