C-H functionalization/C-N bond formation: Copper-catalyzed synthesis of benzimidazoles from amidines

Article abstract of DOI:10.1002/anie.200705420

(Chemical Equation Presented) Copper closes the ring: Benzimidazoles are synthesized from amidines through a copper-catalyzed C-H functionalization/ C-N bond-forming process. The method tolerates a broad range of functional groups and provides the benzimidazoles in up to 89% yield. Best results are obtained by using 15 mol% Cu(OAc)₂, 2-5 equivalents of HOAc as additive, and oxygen as the stoichiometric reoxidant (see scheme).

Full text of DOI:10.1002/anie.200705420

Communications
DOI: 10.1002/anie.200705420
À
C H Functionalization
À
À
C H Functionalization/C N Bond Formation: Copper-Catalyzed
Synthesis of Benzimidazoles from Amidines**
Gordon Brasche and Stephen L. Buchwald*
An immense effort has been made to develop efficient
Cu(OAc)2 in DMSO at 1008C under an oxygen atmosphere
produced 2 after 18 h in 19% yield with a low conversion of 1
(Table 1, entry 1). Using 5 equivalents of pyridine or triethyl-
amine as a basic additive, the conversion was enhanced, but
À
strategies for the direct functionalization of C H bonds using
transition-metal catalysis.[1] For the most part, ruthenium-,[2]
rhodium-,[3] and palladium-based[4] catalysts have been
À
À
applied to effect either C C or C Het (Het = heteroatom)
À
bond formation by replacement of a C H bond. Few
Table 1: Optimization of the initial screening results.
examples have been described that employ copper as
catalysts,[5] which is particular attractive due to its low cost
and low toxicity. Herein, we disclose a new Cu(OAc)2-
catalyzed synthesis of benzimidazoles from amidines through
À
À
a C H functionalization/C N bond-forming process that uses
oxygen as the oxidant and generates water as the only direct
waste product.
Entry Cu catalyst
Additive
(5 equiv)
Gas atm Conv. Yield
[%]
[%][a]
Recently, we reported that carbazoles could be formed by
1
2
3
4
5
6
7
8
9
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
CuCl2
CuSO4
Cu(isobutyrate)2 HOAc
Cu(OAc)2
Cu(OAc)2
Cu(OAc)2
-
O2
O2
O2
O2
O2
O2
O2
O2
37
57
72
96
29
85
92
39
89
70
19
25
23
intramolecular C H bond functionalization.[6] We were eager
À
pyridine
NEt3
HOAc
HOAc
HOAc
to extend our approach to the synthesis of other classes of
heterocycles, such as benzimidazoles, which are of consider-
able importance in medicinal chemistry.[7] Brainꢀs palladium-
catalyzed N-arylation of (ortho-bromophenyl)amidines to
give benzimidazoles[8] inspired us to use amidines as our
starting materials. The cyclization of amidines by oxidative
means had been reported earlier. However, the use of
74 (63)[b]
15
68
71
19
68
formic acid
propionic acid O2
HOAc
10
air
44
[9]
stoichiometric amounts of reagents, such as Pb(OAc)4 and
[a] GCyield vs. internal standard; reactions conducted on a 0.2 mmol
scale. [b] Yield of isolated compounds; average of two runs on a
1.0 mmol scale.
iodine(III) compounds,[10] and a rather narrow functional-
group scope—only examples with Me, Cl, and Br substituents
were reported—are major disadvantages. NaOCl is known to
promote the cyclization as well via N-chlorinated amidines.[11]
Although the low cost of NaOCl is an appealing feature, low
to moderate yields were obtained for amidines bearing
functional groups, and as mentioned in one report,[11b]
chlorinated side products might be anticipated when this
method is applied. Altogether, these drawbacks prompted us
to reinvestigate this transformation.
the yield of 2 was not greatly affected (Table 1, entries 2, 3). A
satisfactory result was obtained when 5 equivalents of acetic
acid were used in lieu of a base additive; a 74% yield of
benzimidazole 2 and almost complete conversion of 1 were
observed (Table 1, entry 4). We subsequently tested other
copper(II) salts and acids (Table 1, entries 5–9). Only Cu-
(isobutyrate)2 and propionic acid provided comparable
results. Catalytic activity was also found under an air
atmosphere, though the formation of 2 was slower (Table 1,
entry 10).
We started our study by examining the conversion of N-
phenylbenzamidine (1) into 2-phenylbenzimidazole (2). After
an initial screen of palladium and copper catalysts, solvents,
and reaction temperatures, we found that the use of 15 mol%
Using the optimized conditions, we next explored the
scope and generality of the process. Amidines were prepared
through the addition of an aniline to a carbonitrile derivative
following known procedures.[12] We were surprised to find that
functionalized amidines derived from aryl nitriles without
ortho substituents showed little conversion into the desired
benzimidazole and mainly the formation of decomposition
products which seemed to inhibit the catalytic cycle. How-
ever, clean formation of substituted 2-phenylbenzimidazoles
was observed when the amidine was derived from a 2-
substituted aryl nitrile.[13] As shown in Table 2, a variety of
ortho substituents, such as Me, CF3, OMe, Cl, or TBS can be
used. In addition, several functional groups including halo-
[*] Dr. G. Brasche, Prof. Dr. S. L. Buchwald
Department of Chemistry, Room 18-490
Massachusetts Institute of Technology
77 Massachusetts Avenue, Cambridge, MA 02139 (USA)
Fax: (+1)617-253-3297
E-mail: sbuchwal@mit.edu
Homepage: http://mit.edu/chemistry/buchwald
[**] We thank the NIH (Grant GM-058160) for support. We thank
Amgen, Merck, and Boehringer Ingelheim for additional unre-
stricted funds. G.B. thanks the German Academic Exchange Service
(DAAD) for a fellowship.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
1932
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1932 –1934

Angewandte
Chemie
Table 2: Synthesis of substituted 2-arylbenzimidazoles.
The method can be extended to the preparation of N-
methylated 2-phenylbenzimidazoles, a feature which was not
reported for the older methods (Table 3).[9–11] Using our
approach, the reaction conditions had to be modified only
Entry Substrate
Product
R1
Yield
[%][a]
Table 3: Synthesis of substituted N-methyl-2-arylbenzimidazoles.
1
2
3
4
5
H
OMe
F
Cl
Br
89
70
86
89
89
Entry
R1
R2
R3
Yield [%][a]
1
2
3
4
H
H
H
Br
H
H
Br
H
Me
C
Cl
C
68
l
6
7
8
9
H
87
69
54
OMe
74[b]
84
CO2tBu 88[c,d]
[b]
l
I
89
[a] Yield of isolated compounds; average of two runs on a 1.0 mmol
scale. [b] 5:1 regioselectivity (HPLC); yield corresponds to the major
isomer shown above.
10
11
H
Cl
68
75
slightly with respect to the amount of acetic acid and reaction
time to obtain high yields. Further investigations also
revealed that 2-alkylated benzimidazoles can be obtained
(Scheme 1). However, compared to the older methods[9–11]
12
13
81
80[e]
14
15
H
Br
85
88
Scheme 1. Synthesis of substituted 2-tert-butylbenzimidazoles. [a] Yield
16
89
of isolated compounds; average of two runs on a 1.0 mmol scale.
our method is limited to amidines bearing a bulky tert-butyl
group. In attempts to cyclize amidines with a smaller ethyl,
isopropyl, or benzyl substituent, only decomposition of the
starting material was observed.
17
80[c,f]
We currently are unsure of the mechanistic course of the
benzimidazole-forming process. Scheme 2outlines possible
pathways suggested by literature precedent of related pro-
cesses. The reaction of 1 with Cu(OAc)2 presumably leads
[a] Yield of isolated compounds; average of two runs on a 1.0 mmol
scale. [b] Reaction also conducted with the meta substituted amidine to
give mainly the same product; 74% yield of isolated compound; 19:1
regioselectivity (GC). [c] Reaction time 36 h. [d] Reaction also conducted
with the meta substituted amidine to give the same product; 69% yield
of isolated compound; no regioisomer detected. [e] Reaction conducted
on a 0.5 mmol scale instead of 1.0 mmol. [f] 14:1 regioselectivity
(HPLC).
gens and electron-donating or electron-withdrawing substitu-
ents are tolerated well on the N-aryl ring of the amidine. The
ability to incorporate the whole range of halogen substituents
makes this method particularly appealing, since these sub-
stituents can be used for further synthetic manipulations. As
for substitution patterns, high yields were obtained when the
substituents are in the 5- and 6-positions of the benzimidazole.
Only low conversions were observed for cyclizations leading
to 4-substituted or 4,6-disubstituted products.
Scheme 2. Possible reaction pathways for the conversion of 1 into 2.
Angew. Chem. Int. Ed. 2008, 47, 1932 –1934ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1933

Communications
first to a Cu–N adduct 3[10] with copper either in oxidation
[3] For selected examples, see: a) B. J. Stokes, H. Dong, B. E. Leslie,
A. L. Pumphrey, T. G. Driver, J. Am. Chem. Soc. 2007, 129,
7500 – 7501; b) J. C. Lewis, R. G. Bergman, J. A. Ellman, J. Am.
Chem. Soc. 2007, 129, 5332– 5333; c) K. Williams Fiori, J. Du
Bois, J. Am. Chem. Soc. 2007, 129, 562– 568; d) P. R. Ravise-
khara, H. M. L. Davies, Org. Lett. 2006, 8, 5013 – 5016.
[4] For selected examples, see: a) D. R. Stuart, E. Villemure, K.
Fagnou, J. Am. Chem. Soc. 2007, 129, 12072 – 12073; b) K. L.
Hull, M. S. Sanford, J. Am. Chem. Soc. 2007, 129, 11904 – 11905;
c) S. Yang, B. Li, X. Wan, Z. Shi, J. Am. Chem. Soc. 2007, 129,
6066 – 6067; d) K. Inamoto, T. Saito, M. Katsuno, T. Sakamoto,
K. Hiroya, Org. Lett. 2007, 9, 2931 – 2934; e) J. H. Delcamp,
M. C. White, J. Am. Chem. Soc. 2006, 128, 15076 – 15077.
[5] a) H.-Q. Do, O. Daugulis, J. Am. Chem. Soc. 2007, 129, 12404 –
12405; b) X. Chen, X.-S. Hao, C. E. Goodhue, J.-Q. Yu, J. Am.
Chem. Soc. 2006, 128, 6790 – 6791; c) Z. Li, S. Bohle, C.-J. Li,
Proc. Natl. Acad. Sci. USA 2006, 103, 8928 – 8933; d) for an
approach using a stoichiometric amount of copper, see: T.
Uemura, S. Imoto, N. Chatani, Chem. Lett. 2006, 35, 842– 843.
[6] W. C. P. Tsang, N. Zheng, S. L. Buchwald, J. Am. Chem. Soc.
2005, 127, 14560 – 14561.
[7] For recent reports on pharmaceuticals containing the benzimi-
dazole core, see: a) A. Figueiras, J. M. G. Sarraguꢁa, R. A.
Carvalho, A. A. C. C. Pais, F. J. B. Veiga, Pharm. Res. 2007, 24,
377 – 389; b) G. Ragno, A. Risoli, G. Ioele, M. De Luca, Chem.
Pharm. Bull. 2006, 54, 802– 806; c) M. Boiani, M. Gonzalez,
Mini-Rev. Med. Chem. 2005, 5, 409 – 424.
[8] a) C. T. Brain, J. T. Steer, J. Org. Chem. 2003, 68, 6814 – 6816;
b) C. T. Brain, S. A. Brunton, Tetrahedron Lett. 2002, 43, 1893 –
1895.
state II or III.[14] In pathway A the N-aryl ring attacks the
amidine moiety in a fashion similar to an electrophilic
aromatic substitution to form 4 with concurrent release of a
reduced copper species. Rearomatization then provides 2. In
pathway B the N-phenyl ring attacks the copper center in 3 to
give a metallacycle 5.[15] Product 2 is subsequently formed
through rearomatization and reductive elimination of the
metal. Pathway C involves a copper nitrene 6.[16] A concerted
À
insertion of the nitrogen into a C H bond or an electrocyclic
ring closure and a final [1,3]-shift of a hydrogen would then
lead to 2.[9]
To probe the action of an electrophilic aromatic substi-
tution mechanism, we treated an unsubstituted amidine and
amidines substituted with either an electron-donating (OMe)
or electron-withdrawing group (CO2tBu) in the para or meta
position of the aniline-derived part of 1. We then measured
and compared their conversions into the corresponding
benzimidazoles after 2, 4, 6, and 8 h.[17] The highest reactivity
was observed for the amidine with an OMe in the meta
position, the lowest for the amidine with a CO2tBu group in
the same position. This result is consistent in direction, but not
in magnitude, with the ability of these substituents to stabilize
the carbocation in 4 or 5. A thorough mechanistic study is
needed to unravel the mechanistic intricacies of this process.
À
In conclusion, an efficient copper-catalyzed C H func-
À
tionalization/C N bond-forming approach providing benzi-
[9] S. Chaudhury, A. Debroy, M. P. Mahajan, Can. J. Chem. 1982, 60,
1122 – 1126.
[10] a) C. A. Ramsden, H. L. Rose, J. Chem. Soc. Perkin Trans. 1
1997, 2319 – 2327; b) C. A. Ramsden, H. L. Rose, J. Chem. Soc.
Perkin Trans. 1 1995, 615 – 617.
[11] a) T. Hisano, M. Ichikawa, K. Tsumoto, M. Tasaki, Chem.
Pharm. Bull. 1982, 30, 2996 – 3004; b) V. J. Grenda, R. E. Jones,
G. Gal, M. Sletzinger, J. Org. Chem. 1965, 30, 259 – 261.
[12] a) B. Singh, J. C. Collins, Chem. Commun. 1971, 498 – 499; b) J. I.
Ogonor, Tetrahedron 1981, 37, 2909 – 2910.
midazoles in good to very good yields from readily available
amidines has been developed. The new method requires only
inexpensive reagents, such as copper(II) acetate, acetic acid,
and oxygen, and tolerates a variety of useful functional
groups.
Received: November 26, 2007
Published online: January 28, 2008
[13] A similar observation in yield improvement by increasing the
steric bulk was made by Bergman and Ellman et al. in a
À
À
Keywords: C H functionalization · C N bond formation ·
copper · heterocycles · homogeneous catalysis
.
À
rhodium-catalyzed C H functionalization. See reference [3b]
for details.
[14] For a discussion of mechanisms involving CuIII intermediates,
see: S. V. Ley, A. W. Thomas, Angew. Chem. 2003, 115, 5558 –
5607; Angew. Chem. Int. Ed. 2003, 42, 5400 – 5449.
[15] For a report suggesting a copper metallacycle as intermediate,
see: K. Yamada, T. Kubo, H. Tokuyama, T. Fukuyama, Synlett
2002, 231 – 234.
À
[1] For recent reviews of C H functionalization using transition
metals, see: a) D. Alberico, M. E. Scott, M. Lautens, Chem. Rev.
2007, 107, 174 – 238; b) K. Godula, D. Sames, Science 2006, 312,
67 – 72; c) F. Kakiuchi, N. Chatani, Adv. Synth. Catal. 2003, 345,
1077 – 1101; d) V. Ritleng, C. Sirlin, M. Pfeffer, Chem. Rev. 2002,
102, 1731 – 1769.
[16] Copper nitrenes have been discussed in the aziridination of
alkenes, see: a) K. M. Gillespie, E. J. Crust, R. J. Deeth, P. Scott,
Chem. Commun. 2001, 785 – 786; b) P. Brandt, M. J. Sꢂdergren,
P. G. Andersson, P.-O. Norrby, J. Am. Chem. Soc. 2000, 122,
8013 – 8020; c) Z. Li, R. W. Quan, E. N. Jacobsen, J. Am. Chem.
Soc. 1995, 117, 5889 – 5890.
[2] For selected examples, see: a) Y. Matsuura, M. Tamura, T. Kochi,
M. Sato, N. Chatani, F. Kakiuchi, J. Am. Chem. Soc. 2007, 129,
9858 – 9859; b) J. M. Murphy, J. D. Lawrence, K. Kawamura, C.
Incarvito, J. F. Hartwig, J. Am. Chem. Soc. 2006, 128, 13684 –
13685; c) L. Ackermann, A. Althammer, R. Born, Angew.
Chem. 2006, 118, 2681 – 2685; Angew. Chem. Int. Ed. 2006, 45,
2619 – 2622.
[17] See Supporting Information for data and a more detailed
discussion.
1934
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 1932 –1934