Selective one-pot synthesis of symmetrical and unsymmetrical di- and triarylamines with a ligandless copper catalytic system

Article abstract of DOI:10.1039/c2cc32252h

The one-pot synthesis of symmetrical or unsymmetrical di- or triarylamines using aryl iodides or bromides and LiNH₂ as ammonia source is reported. This highly selective method is based, for the first time, on a copper-catalyzed system, which does not require the presence of any additional ligand.

Full text of DOI:10.1039/c2cc32252h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 6408–6410
www.rsc.org/chemcomm
COMMUNICATION
Selective one-pot synthesis of symmetrical and unsymmetrical di- and
triarylamines with a ligandless copper catalytic systemw
Anis Tlili, Florian Monnier* and Marc Taillefer*
Received 28th March 2012, Accepted 28th April 2012
DOI: 10.1039/c2cc32252h
The one-pot synthesis of symmetrical or unsymmetrical di- or
triarylamines using aryl iodides or bromides and LiNH2 as
ammonia source is reported. This highly selective method is
based, for the first time, on a copper-catalyzed system, which
does not require the presence of any additional ligand.
In 2007, our group observed that aqueous ammonia (NH3ÁH2O)
could be coupled with aryl halides to selectively afford the
corresponding anilines, using a very simple Cu–diketone catalytic
system.8 We then tried to extend the method via a parametric
study to the synthesis of symmetrical di- and triarylamines.
Unfortunately only non-selective processes were observed, yielding
mixtures of anilines and diarylethers, the latter probably resulting
from the Cu-catalyzed arylation of water.9
Among the variety of amines, di- and triarylamines constitute
a wide family of active compounds in pharmaceutical and
material sciences.1 They have been commonly prepared via
Pd-,2 Ni-3 or Cu-catalyzed4 coupling reactions between aryl
halides and aniline derivatives. Recently, several approaches
employing palladium catalysts have allowed the synthesis of
symmetrical di- or triarylamines from aryl halides as the only
source of arenes in combination with ammonia or ammonia
surrogates. Thus, the preparation of these compounds by
Pd-catalyzed arylation of urea and LiNH2 has been described
by Beletskaya et al.5a and Huang and Buchwald.5b Although
these methods are efficient, their scope is limited to aryl halides
possessing electron-donating groups and is strongly dependent
on the position of the aryl substituents. More recently, Surry
and Buchwald reported a Pd-based catalytic system able to
couple ammonia with aryl bromides or chlorides, to produce
di- or triarylamines using a multistep protocol.6 For the
synthesis of unsymmetrical molecules, the anilines were first
generated from NH3 and aryl halides with the assistance of a
Pd2dba3/(o-biaryl)-di-tert-butylphosphine system. After removing
excess ammonia and solvent, another biaryl-dicyclohexylphosphine
was in turn added, together with a second aryl halide (diarylamine
synthesis), while the further addition of a third aryl halide allowed
the formation of unsymmetrical triarylamines. Although this
method provides an elegant route to di- and triarylamines, it suffers
from drawbacks such as the use of expensive systems based on
palladium salts and of two different elaborated phosphine ligands.
As part of our studies on the copper-catalyzed arylation of
nucleophiles,4a,7–9 we now report a new method allowing the
one-pot selective synthesis of symmetrical or unsymmetrical
di- or triarylamines via a very simple ligandless copper-catalyzed
system involving aryl iodides and LiNH2 as ammonia source.
We then turned our attention to lithium amide as a potential
ammonia source and observed that this compound, already
used for the Pd-catalyzed synthesis of anilines10 and symmetrical
diarylamines,5b was a preferable candidate to reach our goal. First,
we selected PhI as the model substrate and a ligandless copper
catalytic system (CuI) for the optimization of the reaction condi-
tions. The preliminary studies showed that with DMF as solvent,
the use of an excess of LiNH2 (5 eq.) affords a poor yield of
diphenylamine 1 (Table 1, entry 1), together with aniline (20%) and
N-phenylformamide PhNHCHO (38%). The presence of the latter
was not really surprising, as it corresponds to a transamination
between an aniline and DMF, a reaction known to be activated
under strongly basic conditions (Scheme 1).11 Decreasing the
Table 1 Selective synthesis of symmetrical di- or triphenylamines
(1 and 11)a
Entry
LiNH2 (n eq.)
Base (x eq.)
1b (%)
11b (%)
1
2
3
4
5
6
7
8
5
2
2
2
2
2
1
0.5
0.5
0.5
0.5
—
—
20c
60d
66
79
92
0
0
0
0
0
0
0
40
63
87
67
K2CO3 (1)
Cs2CO3 (1)
K3PO4 (1)
K3PO4 (1)
K3PO4 (1)
K3PO4 (1)
K3PO4 (2)
K3PO4 (3)
Cs2CO3 (3)
68e, 37f
72
20
10
3
9
10
11
6
a
Performed with 0.1 mmol of CuI (10 mol%), 1 mmol of PhI, n equiv.
b
of LiNH2, x equiv. of base, 2 mL of solvent at 130 1C for 24 h. GC
yield in % determined with 1,3-dimethoxybenzene as an internal
CNRS, UMR 5253, Institut Charles Gerhardt Montpellier, ENSCM,
8, rue de l’Ecole Normale, 34296 Montpellier Cedex 5, France.
E-mail: florian.monnier@enscm.fr, marc.taillefer@enscm.fr
w Electronic supplementary information (ESI) available. See DOI:
10.1039/c2cc32252h
c
standard. Complete conversion of PhI, along with the formation of
d
PhNH2 (20%) and PhNHC(O)H (38%). Formation of PhNHC(O)H
e
f
(16%). Performed at 110 1C. Performed at 90 1C.
c
6408 Chem. Commun., 2012, 48, 6408–6410
This journal is The Royal Society of Chemistry 2012

View Article Online
Scheme 1 Possible pathways towards symmetrical di- and
triphenylamines.
amount of LiNH2 to two equivalents resulted in a significant
increase in the yield of 1 to 60% (Table 1, entry 2), but the
undesired phenylformamide was still observed (16%). Finally, we
were pleased to find that after the addition of 1 equivalent of an
inorganic base (particularly K3PO4), iodobenzene can be quantita-
tively coupled with LiNH2 to selectively afford diphenylamine 1 in
92% yield (entry 5). As a general feature, it should be noted that
the reaction can proceed at a lower temperature (110 1C, entry 6)
and that DMF was the only solvent12 that allowed the reaction to
proceed to completion. In another set of experiments, we again
reduced the amount of LiNH2 to 1 and 0.5 equivalent and
observed a decrease in the yield of diphenylamine 1 to 72% and
20%, respectively (entries 7 and 8). This last case was particularly
interesting because we detected, for the first time, the formation of
a significant amount of triphenylamine 11. Thereafter, we focused
our efforts towards the selective synthesis of this compound and
observed that the addition of one more equivalent of K3PO4 leads
to a 63% yield of 11, while a 10% yield of diphenylamine 1 was still
obtained (entry 9). Finally, we were glad to isolate 87% of 11 by
using 3 equivalents of K3PO4 (entry 10), this base being the best
candidate for achieving the selectivity of this transformation.
In Scheme 1 is represented a plausible general pathway leading
to the observed products. Even though it seems obvious that a
key factor for the success of our system is the use of a well-defined
ratio of LiNH2/inorganic base, the mechanism is not elucidated
and particularly the relationship between experimental conditions
and reactivity/selectivity is not clear.
Scheme 2 Selective one-pot access to symmetrical di- and triarylamines
1–18.
We then focused our research on the synthesis of unsymmetrical
arylated amines, which constitute another stimulating challenge in
terms of selectivity and applications. Under the conditions given in
Scheme 3 involving two different aryl halides, LiNH2 and the
ligandless copper system allowing the synthesis of symmetrical
diarylamines (Scheme 2, A), we were pleased to obtain the various
On the basis of these results, we developed two highly
selective conditions for the synthesis of various symmetrical
diarylamines 1–10 and symmetrical triarylamines 11–18
(Scheme 2). Under condition A, 10 mol% of CuI efficiently
promotes cross-coupling between LiNH2 and aryl iodides
bearing electron-withdrawing groups (Scheme 2, 2–5) to afford
the corresponding diarylamines in good to excellent isolated
yields (65 to 82%). Similar results were obtained starting from
aryl iodides substituted by electron-donating groups in para or
meta-positions (Scheme 2, 6–9, yields 77–94%). On the other
hand, we have shown that under condition B, the corres-
ponding triarylamines are obtained in good yields with high
selectivities. The reactions well tolerate substituents on the
starting aryl iodides (Scheme 2, 11–16, yields 78–98%), while
the presence of an additional ligand (DMEDA) is required to
obtain a good yield (86%) in the particular case of p-trifluoro-
methyliodobenzene. It is noteworthy that the method was also
efficient for the synthesis of secondary dipyridylamine 10 and
tertiary tripyridylamines 17–18, obtained from 2-bromo-6-
methyl pyridine (Scheme 2, 10 and 18) and 2-bromopyridine
(Scheme 2, 17). The family of polypyridylamines has found a
wide variety of applications as ligands in coordination chemistry
and in organometallic catalysis.13
Scheme 3 Selective one-pot access to unsymmetrical diarylamines 19–30.
Chem. Commun., 2012, 48, 6408–6410 6409
c
This journal is The Royal Society of Chemistry 2012

View Article Online
unsymmetrical diarylamines 19–30 in good to excellent yields
with a high selectivity.
We realized that the use of a small excess of the first introduced
aryl iodide R–C6H4I was needed to almost fully eliminate the
formation of symmetrical diarylamines, which can result from each
of the two aromatic partners. Another key element for this easy-to-
operate protocol was the in situ introduction of the second (hetero-)
aryl iodide or bromide after 6 h of the reaction without any
modification or additional work-up of the reaction mixture. It is
noteworthy to mention that this simple catalytic system is tolerant
towards both electron-donating (4-CH3, 3-OMe) or withdrawing
(F, Br, CF3, CN, COMe) substituents on the two aryl halide
derivatives, the expected products being always obtained in good to
excellent yields (Scheme 3, 19–30). Interestingly, the latter
could be sometimes optimized by changing the introduction
order of the two aromatic partners (Scheme 3, 24–26). Finally,
of particular interest is the possibility to build various mixed
unsymmetrical aryl–heteroaryl(2-pyridyl)amines (Scheme 3, 27–30)
in almost quantitative yields using this method.
Scheme 5 One-pot access to unsymmetrical triarylamine 35.
for the first time, on a Cu-catalyzed system, which does not require
the presence of any additional ligand. Thus, the low cost, the low
toxicity and the simplicity of this catalytic system render the
method competitive with comparable Pd-based protocols, which
require the presence of sophisticated ligands. Work is in progress
to broaden the application field of the method, particularly for the
synthesis of totally unsymmetrical triarylamines.
Notes and references
1 (a) J. R. Kenny, J. L. Maggs, X. Meng, D. Sinnott, S. E. Clarke,
B. K. Park and A. V. Stachulski, J. Med. Chem., 2004, 47, 2816;
(b) R. Kisselev and M. Thelakkat, Macromolecules, 2004, 37, 8951.
2 (a) S. L. Buchwald, C. Mauger, G. Mignani and U. Scholz, Adv.
Synth. Catal., 2006, 348, 23; (b) M. Kienle, S. R. Dubbaka,
K. Brade and P. Knochel, Eur. J. Org. Chem., 2007, 4166.
3 (a) J. P. Wolfe and S. L. Buchwald, J. Am. Chem. Soc., 1997,
119, 6054; (b) B. H. Lipshutz and H. Ueda, Angew. Chem., Int. Ed.,
2000, 39, 4492; (c) C. Desmarets, R. Schneider and Y. Fort, J. Org.
Chem., 2002, 67, 3029.
4 (a) F. Monnier and M. Taillefer, Angew. Chem., Int. Ed., 2009,
48, 6954; (b) G. Evano, N. Blanchard and M. Toumi, Chem. Rev.,
2008, 108, 3054.
5 (a) G. A. Artamkina, A. G. Sergeev, M. M. Stern and I. P. Beletskaya,
Synlett, 2006, 235; (b) X. Huang and S. L. Buchwald, Org. Lett., 2001,
3, 3417.
6 D. S. Surry and S. L. Buchwald, J. Am. Chem. Soc., 2007, 129, 10354.
7 (a) M. Taillefer, H. J. Cristau, P. P. Cellier and J. F. Spindler, Patent
Fr 16547 (WO 03053885), 2001; (b) M. Taillefer, H. J. Cristau,
P. P. Cellier and J. F. Spindler, Chem.–Eur. J., 2004, 10, 5607;
(c) M. Taillefer, N. Xia and A. Ouali, Angew. Chem., Int. Ed., 2007,
46, 934; (d) H. Kaddouri, V. Vicente, A. Ouali, F. Ouazzani and
M. Taillefer, Angew. Chem., Int. Ed., 2009, 48, 333; (e) S. Benyahya,
F. Monnier, M. Wong Chi Men, C. Bied, F. Ouazzani and
M. Taillefer, Green Chem., 2009, 11, 1121; (f) E. Racine,
F. Monnier, J.-P. Vors and M. Taillefer, Org. Lett., 2011, 13, 2818.
8 (a) M. Taillefer and N. Xia, Patent Fr 06827 (WO 2008 050366),
2007; M. Taillefer and N. Xia, Angew. Chem., Int. Ed., 2009, 48, 337.
For other examples (non-exhaustive list) see: (b) J. Kim and
S. Chang, Chem. Commun., 2008, 3052; (c) F. Wu and C. Darcel,
Eur. J. Org. Chem., 2009, 4753 and 2009, 5677; (d) X.-F. Zeng,
W. Huang, Y. Qiu and S. Jiang, Org. Biomol. Chem., 2011, 9, 8224.
9 (a) M. Taillefer, F. Monnier, A. Tlili and N. Xia, Patent Fr 07767
(WO 2010/142913), 2009; (b) A. Tlili, N. Xia, F. Monnier and
M. Taillefer, Angew. Chem., Int. Ed., 2009, 48, 8725; (c) A. Tlili,
F. Monnier and M. Taillefer, Chem.–Eur. J., 2010, 16, 12996.
10 Q. Shen and J. F. Hartwig, J. Am. Chem. Soc., 2006, 128, 10028.
11 Z. Mucsi, G. A. Chass and I. G. Csizmadia, J. Phys. Chem. B,
2008, 112, 7885.
Another challenging task in the synthesis of triarylamines
was to devise a method for the one-pot synthesis of tertiary
amines bearing at least two different aromatic cycles. Among
them, we chose to consider the case of aryldipyridylamines
that are, as outlined above,13 ligands with high potential in
transition metal coordination chemistry and in catalysis.
As a result of various experimentations, it proved possible
to synthesize these molecules from aryl iodides and 2-bromo-
pyridine in excellent yields (Scheme 4, 31–34). The catalytic
system is quite similar to the one that allowed the synthesis of
unsymmetrical diarylamines, the difference lying in the use of
one additional equivalent of K3PO4, presumably to facilitate
the deprotonation of the diarylamine intermediate. It is worthy
of note that this one-pot ligandless copper-catalyzed system is
compatible with electron-donating or withdrawing groups
present on the aryl iodide (Scheme 4).
Finally, we have tried to synthesize, in one step, an even more
challenging triarylamine bearing three different aromatic groups.
By using the former catalytic system, in which a slight excess of
iodobenzene was reacted with 2-bromopyridine and 4-iodobenzoni-
trile, we could conduct the direct one-pot synthesis of tertiary
phenyl(4-cyanophenyl)(pyridin-2-yl)amine 35, obtained in 51% yield
(42% isolated) (Scheme 5). Studies are in progress to render this
method more general and to rationalise the selectivity obtained.14
In summary, we have disclosed a versatile method allowing
the one-pot synthesis of various symmetrical or unsymmetrical
di- or triarylamines using a simple ammonia source (LiNH2) and
aryl halides. This controlled and highly selective process is based,
12 In methylisobutylketone (MIBK) 1 is obtained in 21% yield (0% in
DMSO, H2O or PhMe). Ethanol led to the formation of EtOPh
(91%).
13 (a) N. M. Shavaleev, A. Barbieri, Z. R. Bell, M. D. Ward and
F. Barigelletti, New J. Chem., 2004, 28, 398; (b) A. J. Hickman,
J. M. Villalobos and M. S. Sanford, Organometallics, 2009, 28, 5316;
(c) K. J. Jang, G. Y. Yeo, T. S. Cho, G. H. Eon, C. Kim and
S. K. Kim, Biophys. Chem., 2010, 148, 138; (d) C. Romain, S. Gaillard,
M. K. Elmkaddem, L. Toupet, C. Fischmeister, C. M. Thomas and
J. L. Renaud, Organometallics, 2010, 29, 1992.
14 Unreacted iodobenzonitrile: 45%. Other traces of side-products
are detected: PhNPy2 (11%) and NPy3 (7%).
Scheme 4 New one-pot access to triarylamines aryldipyridylamines 31–34.
6410 Chem. Commun., 2012, 48, 6408–6410
c
This journal is The Royal Society of Chemistry 2012