Pd(0)-catalyzed diarylation of sp3 C-H bond in (2-azaaryl)methanes

Article abstract of DOI:10.1021/ol200345a

A highly efficient and selective palladium-catalyzed diarylation of (2-azaaryl)methanes at the methyl group is described. Aryl chlorides proved reactive enough. A palladium η³-azaallyl intermediate has been identified on the basis of DFT studies.

Full text of DOI:10.1021/ol200345a

ORGANIC
LETTERS
Pd(0)-Catalyzed Diarylation of sp3 C-H
Bond in (2-Azaaryl)methanes
2011
Vol. 13, No. 8
1968–1971
Guoyong Song, Yan Su, Xue Gong, Keli Han, and Xingwei Li*
Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Zhongshan Road,
Dalian 116023, China
xwli@dicp.ac.cn
Received February 7, 2011
ABSTRACT
A highly efficient and selective palladium-catalyzed diarylation of (2-azaaryl)methanes at the methyl group is described. Aryl chlorides proved
reactive enough. A palladium η3-azaallyl intermediate has been identified on the basis of DFT studies.
Synthetic methods that utilize metal-catalyzed activa-
tion and subsequent functionalization of sp2 and sp3 C-H
bonds have emerged as powerful tools to directly install
important functional groups to construct complex struc-
tures.1 In particular, palladium catalysts have been widely
used to effectively achieve the functionalization of the
C-H bonds of arenes and heteroarenes in both redox
and redox-neutralsettings.2 These syntheticstrategies have
played increasingly important roles in natural product
synthesis and drug discovery, and they have been exten-
sively reviewed.1b,c,2 In contrast, activation and functiona-
lization of inert sp3 C-H bonds still represent a huge
challenge in organic synthesis.3 Successful examples on
functionalization of methyl groups are rather limited, and
they took advantage of directing groups4 and/or the
relatively high acidity induced by highly withdrawing
groups.5 It is necessary to explore functionalization of
methyl groups activated by other readily installed groups.
Acidity of methyl or other alkyl groups plays an im-
portant role in their arylation reactions. For example,
palladium-catalyzed R-arylation of esters, ketones, and
amides has become a powerful synthetic tool for the
construction of C(sp3)-C(sp2) bonds.5a,b,6 Formation of
enolates by deprotonation of R sp3 C-H protons of
carbonyl compounds should facilitate metalation. Much
like in these carbonyl compounds, acidity of methyl pro-
tons in (2-azaaryl)methanes is enhanced by the azaaryl
(1) (a) Colby, D. A.; Bergman, R. G.; Ellman, J. A. Chem. Rev. 2010,
110, 624. (b) Sun, C.-L.; Shi, Z.-J. Chem. Commun. 2010, 46, 677. (c)
Ackermann, L.; Vicente, R.; Kapdi, A. R. Angew. Chem., Int. Ed. 2009,
48, 9792. (d) Kakiuchi, F.; Kochi, T. Synthesis 2008, 3013. (e) Alberico,
D.; Scott, M. E.; Lautens, M. Chem. Rev. 2007, 107, 3013. (f) Satoh, T.;
Miura, M. Top. Organomet. Chem. 2008, 24, 61. (g) Ritleng, V.; Sirlin,
C.; Pfeffer, M. Chem. Rev. 2002, 102, 1731. (h) Kakiuchi, F.; Murai, S.
Acc. Chem. Res. 2002, 35, 826. (i) Dyker, G. Angew. Chem., Int. Ed. 1999,
38, 1698.
(2) (a) Lyons, T. W.; Sanford, M. S. Chem. Rev. 2010, 110, 1147. (b)
Chen, X.; Engle, K. M.; Wang, D. H.; Yu, J.-Q. Angew. Chem., Int. Ed.
2009, 48, 5094. (c) Xu, L.-M.; Yang, Z.; Shi, Z.-J. Chem. Soc. Rev. 2010,
39, 712.
(3) For a review on sp3 C-H activation, see: (a) Jazzar, R.; Hitce, J.;
Renaudat, A.; Sofack-Kreutzer, J.; Baudoin, O. Chem.;Eur. J. 2010,
16, 2654. (b) Yu, J.-Q.; Giri, R.; Chen, X. Org. Biomol. Chem. 2006, 4,
4041.
(4) (a) 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. (b) Wasa,
M.; Engle, K. M.; Yu, J.-Q. J. Am. Chem. Soc. 2009, 131, 9886. (c)
Lafrance, M.; Gorelsky, S. I.; Fagnou, K. J. Am. Chem. Soc. 2007, 129,
14570. (d) Chaumontet, M.; Piccardi, R.; Audic, N.; Hitce, J.; Peglion,
J.-L.; Clot, E.; Baudoin, O. J. Am. Chem. Soc. 2008, 130, 15157. (e)
Desai, L. V.; Hull, K. L.; Sanford, M. S. J. Am. Chem. Soc. 2004, 126,
9542. (f) Zaitsev, V. G.; Shabashov, D.; Daugulis, O. J. Am. Chem. Soc.
2005, 127, 13154. (g) Giri, R.; Chen, X.; Yu, J.-Q. Angew. Chem., Int. Ed.
2005, 44, 2112. (h) Yang, S.; Li, B.; Wan, X.; Shi, Z. J. Am. Chem. Soc.
2007, 129, 6066. (i) Pastine, S. J.; Gribkov, D. V.; Sames, D. J. Am.
Chem. Soc. 2006, 128, 14220. (j) Kalyani, D.; Deprez, N. R.; Desai, L. V.;
Sanford, M. S. J. Am. Chem. Soc. 2005, 127, 7330.
(5) (a) Johansson, C. C. C.; Colacot, T. J. Angew. Chem., Int. Ed.
2010, 49, 676. (b) Culkin, D. A.; Hartwig, J. F. Acc. Chem. Res. 2003, 36,
234. (c) Campeau, L.-C.; Schipper, D. J.; Fagnou, K. J. Am. Chem. Soc.
2008, 130, 3266. (d) Guo, Y.; Zhao, X.; Zhang, D.; Murahashi, S.-I.
Angew. Chem., Int. Ed. 2009, 48, 2047. (e) Inoh, J.-I.; Satoh, T.; Pivsa-
Art, S.; Miura, M.; Nomura, M. Tetrahedron Lett. 1998, 39, 4673.
(6) (a) Bellina, F.; Rossi, R. Chem. Rev. 2010, 110, 1082. (b) Lloyd-
Jones, G. C. Angew. Chem., Int. Ed. 2002, 41, 953.
r
10.1021/ol200345a
Published on Web 03/15/2011
2011 American Chemical Society

rings. Although the chemistry of azaallylic anions is well-
known in organic synthesis,7 well-established examples of
metal-catalyzed R-arylation of (2-alkyl)azaarenes are rare.8,9
We reason that both the acidity of these azaallylic protons
and ligating ability of the nitrogen atom should facili-
tate a C-H cleavage process. We now report palladium-
catalyzed efficient diarylation of (2-azaaryl)methanes,
where a palladium η3-azaallyl intermediate has been iden-
tified on the basis of DFT studies, and C-H activation
likely proceeds via metalation-assisted intramolecular
deprotonation.
Table 1. Optimization of Diarylation Conditionsa
entry 2 (equiv)
ligand
base
3ad (%) 4ad (%)
PhBr
1
2
3
4
5
6
7
3.0
1.0
3.0
3.0
3.0
3.0
3.0
PtBu3 HBF4 KOtBub
15
2.3
49
79
8
3
PtBu3 HBF4 KOtBu
Considering the significance of heteroaromatics in drug
discovery and in material studies,10 we have chosen ben-
ziimidazole 1a as a substrate. We initiated our studies on
the coupling between 1a and PhBr using Pd(OAc)2 (5mol%)
and a phosphine as a catalyst in the presence of a base
3
3
3
PtBu3 HBF4 KOtBuc
11
0
PtBu3 HBF4 K2CO3
6
PtBu3 HBF4 Cs2CO3
4
10
91
97
3
S-Phos
PCy3
KOtBu
KOtBu
1.5
0
(PhMe, 130 °C). When PtBu3 HBF4 (10 mol %) was used
------------------------------------------------------------------------------------------
3
together with KOtBu, the coupling reaction occurred but
was strongly influenced by the amount of PhBr and a base
(entries 1-3). Both the monoarylation (minor) and the
diarylation (major) products were detected even though an
access of PhBr was used (entries 1 and 2). A good yield of
the diarylation product 4a could be isolated when 3 equiv
of PhBr was used. Other milder bases (K2CO3 and
Cs2CO3) turned out to be unfavorable for this reaction.
The efficiency of this reaction was further improved when
S-Phos was used. However, in most cases both coupled
products were generated. We noted that essentially no
coupling reaction between 1a and PhBr or PhCl occurred
when chelating phosphines (5 mol %) such as XantPhos and
DPEphos were used (entries 10 and 11 and the Supporting
Information). Gratifyingly, when PCy3 was employed
together with 3 equiv of PhBr, 4a was obtained as the only
product in 97% GC yield (entry 7). Importantly, under
these optimized conditions, less reactive, readily available
chlorobenzenes are efficient coupling partners, and 4a was
isolated as the only product in 95% yield (entries 8 and 9).
With the optimized conditions in hand, we further ex-
plored the scope of ArCl substrates in their coupling with
2-methylbenzimidazoles (Scheme 1). 2-Methylbenziimida-
zoles bearing N-alkyl and -aryl groups readily coupled with
PhCl to give 4a-ad. A broad scope of aryl halides bearing
both electron-donating and -withdrawing groups has been
established in their coupling with 1a. These diarylation
products were isolated in high yield and high selectivity. In
addition, sterically hindered aryl bromides can also be
applied. In particular, when mesityl bromide was used,
monoarylation product 3m was isolated in 81% yield.
8
PhCl
3.0
PCy3
KOtBu
NaOtBu
KOtBu
KOtBu
0
0
0
0
96 (94e)
99 (95e)
9f
10
11
PCy3
Xantphos
DPEphos
0
0
a Reaction conditions: 1a (1 mmol), Pd(OAc)2 (5 mol %), 10 mol %
of monodentate ligand or 5 mol % of bidentate ligand, base (3 equiv),
PhMe (3 mL), 130 °C, 12 h, under N2. b 1.5 equiv of base. c 1.0 equiv of
base. d GC yield using 1,3,5-trimethoxybenzene as a standard. e Isolated
yield. f o-Xylene (3 mL) was used.
Interestingly, although the diaylation product 4mb was also
isolated (12%), the second arylation occurred at the mesityl
methyl group instead of at the 2-methylene position. The
chemoselectivity of the second arylation is likely ascribed to
the steric effect. Heteroaryl chlorides also proved applicable,
and 3-chloropyridine coupled smoothly to give the diaryla-
tion product 4h. In contrast, no coupling occurred for
2-chloropyridine, possibly because it oxidatively adds to
Pd(0) to give an unreactive 2-pryidyl-bridged dinuclear
palladium species.11
A broad scope of (2-azaaryl)methanes can be applied as
given in Scheme 1. Condensed heteroaromatics such as
2-methylbenzoxazole, 2-methylthiazole, and 2-methylqui-
noline all underwent smooth coupling with PhCl, and the
coupled products (4n, 4o, and 4s, respectively) were isolated
in high yield. In addition, 4-methylpyrimidine, 3-methylpyr-
idazine, and 2-methylpyrazine are also suitable substrates. It
is noteworthy that in the case of 2-methylpyrazine a minor
triarylation product 4rb was also isolated (13%) in addition
to the diarylation product (4ra, 81%). When 2-picoline was
subjected to the standard conditions, a less clean reaction
was obtained, and product 4t was isolated in 63% yield.
However, 4-picoline stood in sharp contrast, and essentially
no coupled product (<3%) was detected. Comparisons
between reactivity of 2-picoline (63% yield) and 2-methyl-
quinoline (95% yield) in phenylation reactions seem to
suggest that a methyl group attached to a condensed azaarene
undergoes coupling more readily. Indeed, more pronounced
differences were revealed for the coupling of 1,2-dimethyli-
midazole with PhCl. Arylation occurred exclusively at the
(7) Mangelinckx, S.; Giubellina, N.; De Kimpe, N. Chem. Rev. 2004,
104, 2353.
(8) For catalytic C-C coupling possibly via Pd-azaallylic intermedi-
ates using ArBr, see: (a) Niwa, T.; Yorimitsu, H.; Oshima, K. Org. Lett.
2007, 9, 2373. (b) Burton., P. M.; Morris, J. A. Org. Lett. 2010, 12, 5359.
Stoichiometric functionalization of methyl C-H bonds of 2,6-lutidine by
olefins has been reported. See:(c) Guram, A. S.; Jordan, R. F.; Taylor,
D. F. J. Am. Chem. Soc. 1991, 113, 1833.
(9) (a) Trost, B. M.; Thaisrivongs, D. A. J. Am. Chem. Soc. 2008, 130,
14092. (b) Trost, B. M.; Thaisrivongs, D. A. J. Am. Chem. Soc. 2009,
131, 12056.
(10) Joule, J. A.; Mills, K. Heterocyclic Chemistry; 5th ed.; Wiley: New
York, 2010.
(11) Bertani, R.; Berton, A.; Di Bianca, F.; Crociani, B. J. Organo-
met. Chem. 1986, 303, 283.
Org. Lett., Vol. 13, No. 8, 2011
1969

because the nitrogen atom in 2-methylbenzimidazole and
2-picoline should offer stronger chelation assistance than that
in the competitors. In addition, there is no direct correlation
between the acidity and the reactivity of the heterocyclic
substrate. For example, the more acidic 4-picoline (pKa =
32.212) gave essentially no desired product, while diarylation
product 4t was isolated in 63% yield for the less acidic
2-picoline (pKa = 3412).13 These results indicate that tradi-
tional acidity is not the only factor. We reason that the
adjacent nitrogen atom in the heterocyclic substrate should
play an important role possibly by formation of palladium
η3- or η1-azaallyl intermediates. If η1-azaallyl intermediates
are involved, bidentate phosphines should be efficient ligands
leading to Pd(P-P)Ar(η1-azaallyl) intermediates, from
which subsequent C-C reductive elimination can occur.
Indeed, formation and reductive elimination of related
Pd(Xantphos)(aryl)(2-picolyl) have been recently suggested
in the decarboxylative C-C coupling between potassium
2-(2-pyridyl)acetate and PhBr (eq 3).14 In our case,
essentially no coupling was observed for chelating phos-
phines including Xant-Phos (Table 1), which suggests that
η3-azaallyl intermediates are more likely. The observed
higherreactivity for a methyl group on a fused, condensed
(2-azaaryl)methane is also consistent with this proposal
since the formation of palladium η3-azaallyl intermediate
here resulted in only partial disruption of aromaticity in
the azaarene.15
Scheme 1. Pd-Catalyzed Diarylation of (2-Azaaryl)methanesa
a Reaction conditions: 1 (1 mmol), 2 (3 mmol), Pd(OAc)2 (5 mol %),
PCy3 (10mol%), NaOtBu (3 equiv), and o-xylene (5 mL) under N2, 130 °C,
b Aryl bromide (3 mmol) was used.
12 h, isolated yield.
The KIE of this reaction was experimentally estimated
bycomparing the rates of the reactions of1aand 1a-d3 with
mesityl bromide. A control experiment has shown that
deuterium scrambling occurred among the methyl or
methylene groups in 1a, 1a-d3, and their coupled products
when KOtBu is present, indicating that KIE obtained from
competition reactions is not applicable. Thus, the KIE was
estimated to be 1.3 using 1a and 1a-d3 in separate reactions
5-position, where activation of the sp2 C-H bond took
preference (4u). In addition, when the 4- and 5-positions of
an imidazole are blocked as in 1,3-dimethyl-4,5-diphenyli-
midazole, no reaction occurred.
To gain insight into the mechanism of this coupling
reaction, competition reactions have been carried out. The
competition between 1,2-dimethylbenzimidazole and N-
methylbenzoxazole in equimolar ratio revealed that the
latter reacts preferentially with PhCl (eq 1). In another
competition reaction, an equimolar amount of 2-picoline
and 2-methylquinoline was allowed to react with PhCl
(eq 2). As expected, 2-methylquinoline reacted in predo-
minant preference, and products 4tand 4swereobserved in
a 1:99 ratio (GC). These results seem to suggest that four-
membered palladacyclic intermediates are not likely involved
(12) Fraser, R. R.; Mansour, T. S.; Savard, S. J. Org. Chem. 1985, 50,
3232.
(13) Competition reaction of 2-picoline and 4-picoline in equimolar
ratio with PhCl (3 equiv) under the standard conditions revealed that
only 4t was obtained.
(14) Shang, R.; Yang, Z.-W.; Wang, Y.; Zhang, S.-L.; Liu, L. J. Am.
Chem. Soc. 2010, 132, 14391.
(15) Forexamplesofazaallylcomplexesorintermediates:(a)Murakami,
M.; Hori, S. J. Am. Chem. Soc. 2003, 125, 4720. (b) Roberts, J. S.; Klabunde,
K. J. J. Am. Chem. Soc. 1977, 99, 2509. For examples of enolate complexes:
(c) Panichakul, D.; Su, Y.; Li, Y.; Deng, W.; Zhao, J.; Li, X. Organome-
tallics 2008, 27, 6390. (d) Werner, H.; Mahr, N.; Frenking, G.; Jonas,
V. Organometallics 1995, 14, 619.
1970
Org. Lett., Vol. 13, No. 8, 2011

with a large access of mesityl bromide. This small value
suggests that the cleavage of the methyl C-H bond is not
involved in the rate-determining step.
the η1 to the η3-mode and extrusion of the t-BuOH to give
10 is highly energetically favorable. The trihapticity in 10
˚
follows from the typical lengths of Pd-N (2.329 A) and
˚
two Pd-C bonds (2.470 and 2.147 A). An activation
barrier of 24.8 kcal/mol was revealed for the C-C reduc-
tioneliminationofthisη3-azaallylicspecies, and disruption
of this trihapticity is necessary. Our estimated KIE value of
1.3 seems consistent with the DFT results, and the C-H
cleavage is not involved in the date-determining step. An
alternative C-H cleavage via σ-bond metathesis was also
explored where no nitrogen precoordination is involved.
This σ-bond metathesis process occurs with a signifi-
cantly higher activation barrier (ΔGq = 22.3 kcal/mol) and
is therefore kinetically unfavorable (see the Supporting
Information).17
In summary, we have developed palladium-catalyzed
selective C-H cleavage and cross-coupling of (2-azaaryl)-
methanes with aryl chlorides, and in most cases, diaryla-
tion products were observed. A broad scope of (2-azaaryl)-
methanes has been defined. Theoretical studies indicated
that the C-H activation may proceed via metalation-
assisted intramolecular deprotonation to give a key palla-
dium η3-azaallylic intermediate, and this C-H cleavage
process is not rate-determining. The wide scope, high
selectivity, and the facile reaction conditions should make
this synthetic method applicable to the synthesis of com-
plex structures.
Figure 1. Proposed pathway of (2-azaaryl)methane arylation.
PMe3 was used to reduce the computational cost.
Theoretical studies at the DFT level were carried out to
understand the mechanism of this coupling reaction
(Figure 1). Starting from a model complex cis-Pd(Ph)Cl-
(PMe3)2,16 salt metathesis with KOtBu, affords the tert-
butoxide complex in slight endogonicity. Subsequent sub-
stitution of the phosphine trans to phenyl group by an
incoming 1a should readily occur (ΔG = 2.4 kcal/mol). A
transition state (TS1) was located for the C-H cleavage of
intermediate 8 (ΔGq = 13.9 kcal/mol). In this transition
Acknowledgment. We thank the Dalian Institute of
Chemical Physics, CAS, for generous financial support.
This work is also supported by the NSFC (No. 21072188).
We thank Dr. Bo-Lin Lin (Stanford University) for in-
sightful discussions.
˚
state, the O---H (1.182 A) is nearly formed and the methyl
Supporting Information Available. Experimental pro-
cedures, characterization of compounds, and optimized
geometric data. This material is available free of charge
via the Internet at http://pubs.acs.org.
˚
C---H (1.464 A) is substantially elongated. In addition, no
Pd---C interaction was detected. This pathway of C-H
activation is best described as metalation-assisted intra-
molecular deprotonation via a six-membered ring transition
state. The resulting intermediate 9 can be described as an
η1-azaallyl (enaminate) complex. Subsequent slippage of
(17) Wecannotruleoutthepathwayofdirectdeprotonationfollowedby
transmetalation to give cis-Pd(PMe3)2(Ph)(h1-azaallyl) since the deprotona-
tion of 1a by KOtBu was estimated to be DG = 11.2 kcal/mol on the basis of
DFT studies, so the corresponding DGq might be comparable to that in
coordination-assisted intramolecular deprotonation. However, direct C-C
reductive elimination of cis-Pd(PMe3)2(Ph)(h1-azaallyl), if any, is unlikely
because no coupling occurred when bidentate phosphines were used.
Further studies are underway.
(16) The energy barrier for the oxidative addition of PhCl to Pd-
(PMe3)2 was reported to be 29.3 kcal/mol. See: (a) Lam, K. C.; Marder,
T. B.; Lin, Z. Organometallics 2007, 26, 758. (b) Schoenebeck, F.; Houk,
K. N. J. Am. Chem. Soc. 2010, 132, 2496.
Org. Lett., Vol. 13, No. 8, 2011
1971