Copper-promoted C-N bond cross-coupling with hypervalent aryl siloxanes and room-temperature N-arylation with aryl iodide [5]

Full text of DOI:10.1021/ja001305g

7600
J. Am. Chem. Soc. 2000, 122, 7600-7601
Scheme 1. N-Arylation with Aryl Trimethylsiloxane
Copper-Promoted C-N Bond Cross-Coupling with
Hypervalent Aryl Siloxanes and Room-Temperature
N-Arylation with Aryl Iodide
Patrick Y. S. Lam,* Sophie Deudon, Kristin M. Averill,
Renhua Li, Ming Y. He, Philip DeShong,† and
Charles G. Clark
Table 1. N-Arylation of N-H-Containing Substrates with
Hypervalent Aryl Siloxanes
DuPont Pharmaceuticals Co., Experimental Station
P.O. Box 80500, Wilmington, Delaware 19880-0500
ReceiVed April 14, 2000
The formation of C-N bond via cross-coupling reactions1-3
represents an important addition to the synthetic methodologies
for the preparation of nitrogen-containing compounds in phar-
maceuticals, crop-protection chemicals and material sciences. In
contrast to the powerful C-C bond cross-coupling reactions of
Suzuki4a and Stille,4b a need remains for mild (weak base and
room temperature) and general C-N bond cross-coupling reac-
tions for a wide variety of N-H-containing substrates. In recent
years, Buchwald1a and Hartwig1b have pioneered palladium-
catalyzed C-N cross-couplings of aryl halides with amines,
anilines, mono-nitrogen azoles and carbamates, in general involv-
ing either strong base (t-BuONa) or elevated temperatures.
Arylbismuths1c-d and arylleads1e have been demonstrated to
undergo copper-promoted N-arylation also at elevated tempera-
tures. More recently, the copper-promoted N-arylation with
arylboronic acids for diverse N-H-containing substrates was
discovered by Chan2 and Lam.3 This methodology was further
extended to include, with limited success, arylstannanes.3b In
further pursuit of an optimum arylmetalloid for this versatile
copper-promoted N-arylation reaction, we would like to report
that hypervalent aryl siloxanes are an efficient alternative to
arylboronic acids for C-N bond formation. This new discovery
offers the advantage of performing a one-pot room-temperature
N-arylation in the absence of strong base, starting with aryl iodide,
via the in situ generation of aryl siloxanes. Organosilicon
compounds have recently been shown5-7 to be effective reagents
for C-C bond cross-couplings.
Addition of an equimolar amount8 of tetrabutylammonium
fluoride (TBAF) to phenyl trimethylsiloxane (1) generates a
hypervalent siloxane species 2 (Scheme 1).5 This silicate species
is a very efficient arylating agent for N-H-containing substrates
in the presence of cupric acetate9 at room temperature under
atmospheric air to generate N-arylated cross-coupled product 3.
For example, for benzimidazole 4 in DMF, 83% isolated yield
of 5a was obtained (Table 1). The reaction is very fast with the
rate of the consumption of benzimidazole (90% in 10 min in
methylene chloride) an order of magnitude faster for siloxane than
was observed for boronic acid.3b
† Department of Chemistry and Biochemistry, The University of Maryland,
College Park, MD 20742.
(1) (a) Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem.
Res. 1998, 31, 805-818. (b) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37,
2046-2067. (c) Arnauld, T.; Barton, D. H. R.; Doris, E. Tetrahedron 1997,
53, 4137-4144. (d) Fedorov, A. Y.; Finet, J.-P. Tetrahedron Lett. 1998, 39,
7979-7982. (e) Lopez-Alvarado, P.; Avendano, C.; Menendez, J. C. J. Org.
Chem. 1995, 60, 5678-5682.
Previously when employing arylboronic acids as arylating
agent, it was essential to add base/ligand (either pyridine or
triethylamine, depending on the substrate).2,3 However, we found
that no base/ligand was necessary for N-arylations with aryl
siloxane (Table 1).10
A variety of other N-H-containing substrates can be arylated
(Table 1). 4-tert-Butylaniline 6 gave 72% yield of 7a in CH2-
Cl2.11 N-Ethylbenzimidazolidinone 8 can be N-phenylated to give
58% yield of 9a in CH2Cl2. 4-Phenylpiperidine 10 also undergoes
cross-coupling to give 37% of 11a. Interestly, 2-picolinamide 12
can be N-phenylated (61%) in DMF.12 In terms of the electronic
(2) (a) Chan, D. M. T.; Monaco, K. L.; Wang, R.-P.; Winters, M. P.
Tetrahedron Lett. 1998, 39, 2933-2936. (b) For an improved version of the
related O-arylation see also: Evans, D. A.; Katz, J. L.; West, T. R. Tetrahedron
Lett. 1998, 39, 2937-2940.
(3) (a) Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.; Winters, M.
P.; Chan, D. M. T.; Combs, A. Tetrahedron Lett. 1998, 39, 2937-2940. (b)
Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.; Averill, K.; Chan, D. M.
T.; Combs, A. P. Synlett 2000, 674-676. (c) For solid-phase version of
N-arylation see: Combs, A. P.; Saubern, S.; Rafalski, M.; Lam, P. Y. S.
Tetrahedron Lett. 1999, 40, 1623-1626. (d) Combs, A. P.; Rafalski, M. J.
Combinatorial Chem. 2000, 2, 29-32.
(4) (a) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-2483. (b)
Mitchell, T. N. Synthesis 1992, 803-815.
(5) Mowery, M. E.; DeShong, P. Org. Lett. 1999, 1, 2137-2140.
(6) (a) Denmark, S. E.; Wu, A. Org. Lett. 1999, 1, 1495-1498. (b) Hiyama,
T.; Hatanaka, Y. Pure Appl. Chem. 1994, 66, 1471-1478.
(8) Excess TBAF (50% equivalent excess) did not affect the yield.
(9) Cupric acetate, the best copper salt used previously2,3 in the N-arylation
of arylboronic acid is required. No product was obtained in its absence.
(10) We believe tetra-n-butylammonium fluoride is not the base since in
theory all fluoride is consumed stoichiometrically by the formation of Si-F
bond with aryl trimethylsiloxane.
(7) For transmetalation from silicon to copper see: (a) Nishihara, Y.;
Ikegashira, K.; Hirabayashi, K.; Ando, J.; Mori, A.; Hiyama, T. J. Org. Chem.
2000, 65, 1780-1787 (ref 18, 20 therein). (b) Ito, H.; Hosomi, A. J. Synth.
Org. Chem. 2000, 58, 274. (c) Nishihara, Y.; Ikegashira, K.; Toriyama, F.;
Mori, A.; Hiyama, T. Bull. Chem. Soc. Jpn. 2000, 73, 985-990. (d) Ito, H.;
Ishizuka, T.; Tateiwa, J.; Sonoda, M.; Hosomi, A. J. Am. Chem. Soc. 1998,
120, 11196-11197. (e) Hirabayashi, K.; Nishihara, Y.; Mori, A.; Hiyama, T.
Tetrahedron Lett. 1998, 39, 7893-7896.
(11) In an attempt to increase the yield for tert-butylaniline, phenyl tris-
(2,2,2-trifluoroethyl)siloxane,5 a more reactive siloxane was used, but the yield
remained the same. The use of phenyl triethylsiloxane instead of phenyl
trimethylsiloxane also gave similar yields in the case of benzimidazole.
(12) On the contrary, benzamide gave only 9% N-arylation after heating
at 70 °C. We are currently investigating the significance and utility of this
chelating effect of R-heteroatoms on N-arylation. We thank Dr. Elizabeth
Hauptman of DuPont CR&D for this observation.
10.1021/ja001305g CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/20/2000

Communications to the Editor
J. Am. Chem. Soc., Vol. 122, No. 31, 2000 7601
Scheme 3. One-Pot N-Arylation with Iodobenzene
Scheme 2. Possible Mechanism of N-Arylation with
Hypervalent Aryl Siloxane
Iodobenzene was reacted with trimethoxysilane in the presence
of catalytic Pd2(dba)3 and P(o-tolyl)3 and i-Pr2EtN in DMF15 under
argon at room temperature. After 12 h, benzimidazole, cupric
acetate and TBAF were added and stirred in air until the starting
material is consumed. N-phenylbenzimidazole 5a was obtained
in 40% isolated yield. This accomplished a one-pot N-arylation
of benzimidazole with iodobenzene at room temperature, in
the absence of strong base. 4-Chlorophenyl siloxane 20b and
4-methoxyphenyl siloxane 20c gave 44 and 41% yields, respec-
tively. That the palladium catalyst from the siloxane formation
first step is not involved in the N-arylation second step was
demonstrated by the same yield of N-phenylbenzimidazole
when palladium was removed with Chelex (Bio-Rad) after the
first step.
In summary, we have discovered the copper-promoted C-N
cross-coupling reaction with hypervalent aryl or vinyl siloxane
and a variety of N-H-containing substrates and the extension to
a room-temperature one-pot N-arylation, in the absence of strong
base, with aryl iodide via in situ generation of siloxane. To the
best of our knowledge, this is the first example of room-
temperature N-arylation with aryl iodide in the absence of strong
base.16 The mild condition of the reaction is analogous to that of
amide C-N bond formation and can potentially tolerate most
base-sensitive functional groups. We are currently investigating
the general scope of this reaction, in particular, other N-H-
containing substrates, the use of aryl bromides/chlorides, catalytic
copper17 and applications in generating heterocycle-containing
libraries.
effect of the substituents on aryl siloxane, the yield is in the
following descending order: p-methoxy > phenyl > p-chloro.
In general, DMF is the preferred solvent for 4, 10, and 12. CH2-
Cl2 is the preferred solvent for 6 while 8 works equally well in
either DMF or CH2Cl2.
That the reaction involves a hypervalent siloxane species is
evident by the fact that no 5a was obtained in the absence of
fluoride. The hypervalent siloxane species can also be preformed
prior to the addition of benzimidazole and cupric acetate with no
change in yields. Water does not appear to interfere with the
reaction as the commercial solution of TBAF in THF contains
5% water. Addition of 4 Å molecular seives thus provides no
change in the yield of N-phenylbenzimidazole. This is again in
contrast to the copper-promoted N-arylation using arylboronic
acids where the use of 4 Å molecular sieves is reported to improve
the yield.2b-3
We have also discovered that N-vinylation is possible. For
example, benzimidazole can be vinylated by vinyl trimethyl-
siloxane 14 in CH2Cl2 to give 88% yield of N-vinylbenzimidazole
15. We are currently investigating the scope and utility of this
novel N-vinylation reaction.13
We believe the mechanism (Scheme 2) is similar to that
postulated for N-arylation with arylbismuths1c-d or arylboronic
acids.2,3 Copper(II) acetate is insoluble in CH2Cl2. The first step3b
involves the rapid coordination and dissolution of copper(II)
acetate by heterocycle such as benzimidazole to form heterocycle‚
copper(II) complex 16. The second step involves the trans-
metalation7 of the pentavalent aryl silicate 2 formed by the
addition of fluoride to aryl trimethoxysiloxane 15-7 with 16 to
give heterocycle‚copper(II)‚aryl complex 17. Complex 17 can
undergo reductive elimination to give 19. Alternatively, 17 can
undergo air oxidation or disproportionation to yield the corre-
sponding higher oxidation-state copper(III) complex 18 which can
be more efficiently reductive eliminate to afford 19. A free radical
mechanism is ruled out since the addition of 1,1′-diphenylethylene
has no effect on the reaction.
Acknowledgment. We thank Dr. Paul S. Anderson and Dr. Ruth R.
Wexler for their support of this research.
Note Added in Proof. Replacing the ligand from P(o-tol)3 to
1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene) in the one-pot
reaction increases the yield of 5a from 40 to 53% (We thank
Professor Nolan for a sample of the ligand: Lee, H. M.; Nolan,
S. P. Org. Lett. 2000, 2, 2053-2055).
Supporting Information Available: Experimental details for the
1
general procedure, H spectra (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
JA001305G
The ultimate goal of our investigation is to perform a one-pot
N-arylation of benzimidazole using iodobenzene 20a as the
arylating agent via in situ generation of phenyl trimethylsiloxane.
Masuda had recently reported the palladium-catalyzed formation
of phenyl trimethylsiloxane from 20a and trimethoxysilane 21 at
room temperature.14 We were attracted to the mild condition of
this aryl siloxane formation reaction. The mild siloxane formation
reaction of Masuda appears to complement our mild siloxane
arylation reaction for a combined one-pot reaction (Scheme 3).
(15) Preliminary investigation of solvent preferences for the one-pot
N-arylation is as follow: DMF, NMP > DMAC . EtOAc, dioxane, THF.
CH2Cl2 gave no product.
(16) In general, N-arylation using Buchwald and Hartwig chemistry requires
either elevated temperature or strong base (Nat-BuO). During the course of
this work, we became aware that some room-temperature N-arylation of
Buchwald/Hartwig chemistry is possible, although still in the presence of stong
base Na t-BuO. (a) Wolfe, J. P.; Buchwald, S. L. Angew. Chem., Int. Ed.
1999, 38, 2413-2416. (b) Hartwig, J. P.; Kawatsura, M.; Hauck, S. I.;
Shaughnessy, K. H.; Alcazar, L. M. J. Org. Chem. 1999, 64, 5575-5580.
(17) Modifying Collman’s recent procedure, we found that 0.1 equiv of
[Cu(OH)‚TMEDA]2Cl2 catalyzes the N-phenylation of benzimidazole with
hypervalent phenyl trimethoxysiloxane (69% yield) at 50 °C in DMF (Collman,
J. P.; Zhong, M. Org. Lett. 2000, 2, 1233-1236).
(13) N-vinylation has not been reported by Buchwald/Hartwig.1a-b
(14) Murata, M.; Suzuki, K.; Watanabe, S.; Masuda, Y. J. Org. Chem.
1997, 62, 8569-8571.