Formation of aryl-nitrogen bonds using a soluble copper(I) catalyst

Article abstract of DOI:10.1016/S0040-4039(01)00888-7

We report a synthetic protocol for the synthesis of functionalized diaryl- and triarylamines under mild conditions, using a soluble, air-stable copper(I) complex, Cu(PPh₃)₃Br, as the catalyst and cesium carbonate as the base. Using this protocol, we were able to synthesize a tri-ortho-ester functionalized triphenylamine, which had eluded us using the palladium chemistry.

Full text of DOI:10.1016/S0040-4039(01)00888-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 4791–4793
Formation of arylꢀnitrogen bonds using a soluble copper(I)
catalyst
Rattan Gujadhur, D. Venkataraman* and Jeremy T. Kintigh
Department of Chemistry, University of Massachusetts, Amherst, MA 01003, USA
Received 27 March 2001; revised 23 May 2001; accepted 24 May 2001
Abstract—We report a synthetic protocol for the synthesis of functionalized diaryl- and triarylamines under mild conditions, using
a soluble, air-stable copper(I) complex, Cu(PPh3)3Br, as the catalyst and cesium carbonate as the base. Using this protocol, we
were able to synthesize a tri-ortho-ester functionalized triphenylamine, which had eluded us using the palladium chemistry.
© 2001 Elsevier Science Ltd. All rights reserved.
Arylamines are prevalent in compounds that are of
biological, pharmaceutical and materials interest.1–3
Their widespread importance has led to the develop-
ment of many synthetic methods for the formation of
arylꢀnitrogen bonds.1,4 Amongst them, the classical
copper-mediated Ullmann coupling and the recently
developed palladium(0)-catalyzed aryl coupling are the
more commonly used methods.5–12 The Ullmann cou-
pling often requires high temperatures (ꢀ200°C) and
the use of copper salts in greater than stoichiometric
amounts. The reaction is also very sensitive to the
substitution on the aryl halide. Due to these limitations,
copper salts have largely been supplanted by palladium
catalysts. However, it is still the reaction of choice for
the synthesis of aryl amines in large and manufacturing
scales.3
size a triphenylamine that was functionalized at the
ortho positions with ester groups.13 This inability of the
palladium(0)-based catalysts coupled with well-known
limitations of classical Ullmann reactions provided us
the impetus to investigate the possibility of developing
copper-mediated reactions, which can occur at milder
temperatures and are of broad scope. In addition, there
is an economic attractiveness for using copper over
noble metals such as palladium.14
Based on our recent success in using Cu(PPh3)3Br (1) as
a catalyst for the formation of diaryl ethers,15 we chose
to explore the efficacy of this soluble copper(I) complex
in the formation of arylꢀnitrogen bonds. In this report,
we extend the utility of this catalyst for the synthesis of
various functionalized diaryl- and triarylamines under
milder reaction conditions (Scheme 1).
We became interested in the copper-mediated reactions
because of a limitation that we encountered with palla-
dium catalysts. In spite of using the most active ligands
to date with palladium(0),10 we were unable to synthe-
Complex 1 can easily be prepared from CuBr2 and
triphenylphosphine.† It is soluble in organic solvents
such as THF, dichloromethane, acetonitrile, chloro-
Cu(PPh3)3Br
(20 mol%)
I
R1
R3
R1
+
R3
R2
N
Cs2CO3, toluene
110 ˚C-120 ˚C
N
H
R2
Scheme 1.
* Corresponding author. E-mail: dv@chem.umass.edu
† In an Erlenmeyer flask equipped with a Teflon stir bar, methanol (100 mL) was heated to boiling and triphenylphosphine (Acros, 6 g, 22.4 mmol)
was slowly added to the stirring methanol. After the complete dissolution of triphenylphosphine, CuBr2 (Acros, 99+%, 1.24 g, 5.27 mmol) was
added as a solid, in portions. No special precautions were taken for the exclusion of air. Upon addition of the copper bromide, a white
precipitate was formed. After the completion of the addition, the contents were stirred for 10 min and the flask was allowed to cool to ambient
temperature. The reaction mixture was then filtered through a Buchner funnel and the white residue was washed repeatedly with ethanol and
then with diethyl ether. The resultant white solid was dried under dynamic vacuum to give 1 (5.73 g, 85% yield, mp 164°C). The cell constants,
contents and the space group are identical to that of the already reported structure of Cu(PPh3)3Br (Cambridge Structural Database Refcode-
FEYVAG).
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00888-7

4792
R. Gujadhur et al. / Tetrahedron Letters 42 (2001) 4791–4793
form, NMP, DMF, DMSO, toluene and benzene.
However, it is not soluble in diethyl ether, hexane,
ethanol or methanol. It can be stored under air for
prolonged times, without any visible color change.
With Cs2CO3 as the base, we were able to use base-sen-
sitive functional groups such as ketones and esters (see
Table 1). Using our protocol, we were also able to
selectively aminate an aryl iodide in the presence of a
bromide (entry 5). Furthermore, we were able synthe-
size the triphenylamine, functionalized at the ortho
positions with ester groups (entry 11)—a molecule that
had eluded the palladium chemistry.
We first examined the propensity of 1 to act as a
catalyst for the formation of triphenylamine from
diphenylamine and bromobenzene or iodobenzene. We
found that the reactions were complete in 24 h (by
TLC) at 120°C when the aryl halide was iodobenzene
and with 20 mol% of 1, cesium carbonate as the base
and toluene as the solvent. Using this protocol, we were
able to synthesize a variety of functionalized triphenyl-
amines in good yields from the readily available
diphenylamine and aryl iodides (entries 1–11, Table 1).‡
However, the reactions were incomplete in 24 h (by
TLC) if NMP was used as the solvent or if the aryl
halide was bromobenzene. Also, we did not observe the
formation of triphenylamine if N,N-dimethylaminopy-
ridine (DMAP), potassium carbonate or sodium
methoxide was used as the base instead of Cs2CO3.
To show the effectiveness of our catalyst, in our syn-
thetic protocol, we replaced our catalyst with 20 mol%
of CuI, CuBr or 20 mol% of CuBr/60 mol% of PPh3. In
each of these reactions, we failed to observe the forma-
tion of triphenylamine. Recently, Goodbrand and Hu
reported that when 1,10-phenanthroline/CuCl used as
the catalyst and KOH as the base, the rate of formation
of arylꢀnitrogen bonds was accelerated and the reac-
tions were complete in 6 h, at 125°C, in toluene.3 To
accommodate base-sensitive functional groups like
esters, we replaced KOH with cesium carbonate in
Goodbrand’s protocol. However, we found that the
reaction of diphenylamine and iodobenzene was incom-
plete even after 15 h.
In general, except for entry 8, electron-deficient aryl
iodides provided better yields of triarylamines than
electron-rich aryl iodides. While we are unable to
explain the low reactivity of methyl-4-iodobenzoate in
entry 8, after various trials, we found that if diphenyl-
amine, 1 and KOtBu (instead of Cs2CO3) were heated
in toluene for 5 min at 120°C, cooled to room tempera-
ture and then methyl-4-iodobenzoate was added and
reheated to 120°C for 24 h, the corresponding triphenyl-
amine was obtained in 62% yield (entry 9).
We also found that 1 can be used as a catalyst for the
formation of diphenylamine from substituted anilines
and aryl iodides (entries 12–15, Table 1). Surprisingly,
we found that these reactions were successful only if the
amine, the catalyst and the base are stirred at 110°C for
few minutes (ꢀ5 min for the formation of an yellow
Table 1. Reactions of arylamines with aryl iodides with 20 mol% of 1 and cesium carbonate
Entry
R1
R2
R3
Time (h)
Temp. (°C)
Isolated yield (%)
1
2
3
4
5
6
7
8
H
H
H
H
H
H
H
H
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
H
o-CH3
o-COOCH3
p-CH3
p-Br
24
24
24
24
24
24
24
24
24
32
32
24
24
24
24
120
120
120
120
120
120
120
120
120
120
175
110
110
110
110
70
52
69
52
54
68
78
25
62a
10
40b
75
88
83
70
p-COCH3
p-NO2
p-COOCH3
p-COOCH3
o-COOCH3
o-COOCH3
H
9
H
Ph
10
11
12
13
14
15
o-COOCH3
o-COOCH3
H
p-CH3
o-COOCH3
o-COOCH3
C6H4-o-COOCH3
C6H4-o-COOCH3
H
H
H
H
H
o-COOCH3
H
a With KOtBu as the base.
b o-Dichlorobenzene was used as a solvent.
‡ In an argon-filled glove box, a Pyrex glass tube (2.5 cm in diameter) equipped with a Teflon stir bar and a rubber septum, was charged with
cesium carbonate (Acros, 1.5 mmol), Cu(PPh3)3Br (20 mol% with respect to the aryl iodide) and diphenylamine (1 mmol). The glass tube was
sealed with a rubber septum and was taken out. The aryl halide (1 mmol) and toluene (15 mL) were injected into the tube through the septum.
The tube was then degassed and back-filled with argon three times using a long needle. The contents were then stirred at 120°C for times
indicated in Table 1. Care was taken to make sure the contents were well stirred. The reaction mixture was then cooled, mixed with 15 mL of
diethyl ether, and filtered to remove any insoluble residues. The combined extracts were dried with anhydrous Na2SO4. The solvent was removed
under reduced pressure and the residue was then purified by flash column chromatography on silica gel to obtain the analytically pure product.

R. Gujadhur et al. / Tetrahedron Letters 42 (2001) 4791–4793
4793
color) before the addition of the aryl iodide. If all the
starting materials were added at room temperature and
heated to 110°C, we observed a black precipitate and
there is no formation of the diphenylamine. All of the
2. Hong, Y.; Senanayake, C. H.; Xiang, T.; Vandenbossche,
C. P.; Tanoury, G. J.; Bakale, R. P.; Wald, S. A. Tetra-
hedron Lett. 1998, 39, 3121–3124.
3. Goodbrand, H. B.; Hu, N.-X. J. Org. Chem. 1999, 64,
670–674.
4. Beller, M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1316–
1317.
5. Lindley, J. Tetrahedron 1984, 40, 1433–1456.
6. Hartwig, J. F. Pure Appl. Chem. 1999, 71, 1417–1423.
7. Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1997,
119, 8232–8245.
8. Hartwig, J. F. Synlett 1996, 4, 329–340.
9. Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999,
576, 125–146.
10. Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J. J.;
Buchwald, S. L. J. Org. Chem. 2000, 65, 1158–1174.
11. Djakovitch, L.; Wagner, M.; Kohler, K. J. Organomet.
Chem. 1999, 592, 225–234.
1
compounds in Table 1 have been characterized by H
and 13C NMR and elemental analyses. Entries 3, 5, 7
and 11 have also been characterized by single crystal
X-ray analyses.
In conclusion, we have shown that a soluble, stable and
an easy-to-prepare copper(I) complex, Cu(PPh3)3Br,
can be used as a catalyst for the formation of
arylꢀnitrogen bonds under milder conditions. This cata-
lyst is selective for aryl iodides and the synthetic proto-
col tolerates various functional groups. We are
currently investigating the mechanistic aspects of this
reaction and the scope of this catalyst in other palla-
dium-catalyzed reactions.
12. Beletskaya, I. P.; Bessmertnykh, A. G.; Guilard, R. Tet-
rahedron Lett. 1999, 40, 6393–6397.
13. Hellwinkel and Melan had reported the synthesis of this
compound in 1971 using traditional Ullmann coupling in
about 20% yield (Chem. Ber. 1971, 104, 1001–1006).
However, in our hands, this reaction was irreproducible.
Hence, we tried the more modern Hartwig–Buchwald
coupling using various ligands such as DPPF, DPPB,
PPh3 including the more recent ligands reported by Buch-
wald (see Ref. 10) and bases such as Cs2CO3, BuLi,
LDA, KHDMS, NaOMe and NaOtBu. From these reac-
tions, we either recovered the starting materials (see entry
10 or 11) or an intramolecular cyclization product in
moderate yields, presumably due to CꢀH activation of
the methyl ester of 2,2%-azanediyl-bis-methylbenzoate.
Van Allen, D.; Field, J. E.; Venkataraman, D.,
manuscript in preparation.
Acknowledgements
The University of Massachusetts, Amherst start-up
funds provided the financial support for this research.
D.V. gratefully acknowledges a Camille and Henry
Dreyfus New Faculty Award. The authors would like
to thank Dr. Greg Dabkowski of the Microanalysis
Laboratory at the UMass-Amherst for elemental analy-
ses. They would also like to thank the University of
Massachusetts Chemistry Department X-ray Structural
Laboratory supported by National Science Foundation
grant CHE-9974648 and Dr. A. Chandrasekaran for
single-crystal X-ray data collection.
14. The price of palladium has risen from $100/ounce in 1995
to $800/ounce in 2001, making it one of the most expen-
sive metals. In comparison, platinum costs $600/ounce
and copper costs $0.05/ounce. For latest prices, see www.
kitco.com.
References
1. Belfield, A. J.; Brown, G. R.; Foubister, A. J. Tetra-
hedron 1999, 55, 11399–11428.
15. Gujadhur, R.; Venkataraman, D. Synth. Commun., in
press.
.