Cobalt-catalyzed C-N bond-forming reaction between N-aromatic 2-chlorides and secondary amines

Article abstract of DOI:10.1002/ejoc.200900597

Secondary amines react with N-aromatic 2-chlorides in the presence of a catalytic amount of cobalt chloride. When DPPP was added as ligand, the yield was further improved, The N-aromatic-containing tertiary amines formed are interesting due to their poten

Full text of DOI:10.1002/ejoc.200900597

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200900597
Cobalt-Catalyzed C–N Bond-Forming Reaction between
N-Aromatic 2-Chlorides and Secondary Amines
Gabriel Toma,*[a] Ken-ichi Fujita,[a,b] and Ryohei Yamaguchi*[a]
Keywords: Cobalt / Cross-coupling / C–N bond formation / Nitrogen heterocycles / Amines
Secondary amines react with N-aromatic 2-chlorides in the
presence of a catalytic amount of cobalt chloride. When
DPPP was added as ligand, the yield was further improved.
The N-aromatic-containing tertiary amines formed are inter-
esting due to their potential biological activity. This work
represents the first cobalt-catalyzed approach to C–N bond
formation involving N-aromatic 2-chlorides and secondary
amines having a certain amount of versatility and functional
group tolerance.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
Representative procedures previously reported include
transition-metal-free microwave procedures[3a] and palla-
dium-catalyzed,[3b] nickel-catalyzed,[3c] and copper-cata-
lyzed[3d] reactions. Reactions performed under microwave
irradiation require a large excess of amine and the yields
are moderate. As a result of the expensiveness of palladium
and complications in the carbon–nitrogen bond-forming
procedures reported for nickel catalysts,[4] the necessity of
new methodologies involving readily available and nontoxic
metals such as cobalt[5] and iron[6] increased. Recent pro-
gresses in cobalt catalysis by the Oshima group,[7] the Gos-
mini group,[8] and other groups[9] mainly regarding cross-
coupling reactions show that cobalt-catalyzed methodolo-
gies are promising tools of an environmentally friendly na-
ture. Because there are some drawbacks consisting in the
use of excess amounts of reducing reagents, involvement of
electrochemical techniques, and the use of large amounts of
expensive ligands, further development in this field is re-
quired. Although a very recent report[5] mentioning a car-
bon–nitrogen bond-forming reaction by using a cobalt cata-
Introduction
One of the methods employed for the formation of car-
bon–heteroatom bonds for the synthesis of important com-
pounds of pharmaceutical and biological interest is the
transitional-metal cross-coupling methodology.[1] Among
carbon–nitrogen bond-forming processes, α-amination of
nitrogen-containing heterocycles is of particular interest,
because those molecules represent building blocks for the
synthesis of biologically active compounds.[2] In Figure 1
some representative 2-pyridine-included bioactive mole-
cules and their purpose are shown. Along with the 2-pyr-
idine moiety, a six-member nitrogen-containing ring is part
of the structure of such molecules. That is why we focused
our attention on synthesizing molecules containing such
six-membered nitrogen-containing rings.
lyst/DMEDA
(N,NЈ-dimethylethylenediamine)
system
showed that aromatic iodides react with a restricted range
of nitrogen nucleophiles, a low turnover number (TON; 1.6
to 2.4) for cobalt chloride was reported in the case of aro-
matic bromides, and the reactivity of heterocyclic halides
Figure 1. Examples of bioactive molecules including one or more was not investigated. Herein, we report the first cobalt-cata-
N-containing heterocycles.
lyzed approach to C–N bond formation involving N-aro-
matic 2-chlorides and secondary amines having a certain
[a] Graduate School of Human and Environmental Studies, Kyoto
amount of versatility and functional group tolerance.
University,
Kyoto 606-8501, Japan
Fax: +81-75-753-6634
E-mail: tomagabi@f06.mbox.media.kyoto-u.ac.jp
Results and Discussion
yama@kagaku.mbox.media.kyoto-u.ac.jp
[b] Graduate School of Global Environmental Studies, Kyoto Uni-
versity,
Kyoto 606-8501, Japan
The reaction was screened for different bases and ligands
in order to optimize the reaction conditions, and the results
are shown in Scheme 1 and Table 1.
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900597.
4586
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 4586–4588

Cobalt-Catalyzed C–N Bond-Forming Reaction
the presence of functional groups affected the yield signifi-
cantly. In the case of an electron-donating group such as
methyl, the yield was low (Table 2, Entry 2). In the case of
electron-withdrawing groups, the yields were good (75 and
60%; Table 2, Entries 3 and 4) and the functional groups
were not altered. Incorporation of a fused benzene ring in-
creased the reactivity compared to 2-chloropyridine; high
yields were obtained for 2-chloroquinoline (88%; Table 2,
Entry 6) and 2-chloroquinoxaline (91%; Table 2, Entry 7)
in the reaction with piperidine.
Scheme 1. Cobalt-catalyzed coupling of 2-chloropyridine with pip-
eridine.
Table 1. Optimization of the cobalt-catalyzed cross-coupling reac-
tion.
Table 2. Reactions of N-aromatic 2-chlorides with piperidine in the
presence of cobalt(II) chloride (10 mol-%) and DPPP (10 mol-%)
at 135 °C in p-xylene.
Entry Base
Ligand
Catalyst
Yield[a] [%]
1
K3PO4
none
none
none
none
CoCl2
CoCl2
CoCl2
CoCl2
Co(acac)2
CoCl2
CoCl2
CoCl2
CoCl2
none
CoCl2
CoCl2
CoCl2
CoCl2
CoCl2
29
59
33
30
0
2
3
4
5
6
7
8
9
K2CO3
Na2CO3
Li2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
none
-proline
TMEDA
DPPE
DPPP
DPPP
DPPB
PPh3
DPPP
none
DPPP
48
37
56
80
trace
53
10
11
12
13[c]
14[c]
15[d]
61 (65[b]
)
50
37
5
[a] GC yield by using cumene as internal standard. [b] 20 mol-%
ligand was used. [c] 5 mol-% catalyst was used. [d] 1 mol-% catalyst
was used, 5 mmol-scale reaction.
At first, we found that cobalt(II) chloride gave promising
results and, therefore, chose to fix it during the optimization
of the reaction conditions. When cobalt(II) acetylacetonate
was used (Table 1, Entry 5), product formation was not ob-
served by GC analysis. Addition of potassium carbonate
showed the highest yield among the bases screened (Table 1,
Entries 1–4) and therefore it was further used for ligand
screening (Table 1, Entry 2).
Amine ligands (Table 1, Entries 6 and 7) showed a sup-
pressing effect on the reaction, suggesting that the nitrogen
atom of 2-chloropyridines may act as a ligand in the reac-
tion and, therefore, amine ligands could decrease the yield.
Phosphane ligands either improved the yield (Table 1,
Entries 9 and 12) or did not improve the yield (Table 1, En-
tries 8 and 11). Triphenylphosphane gave a slightly better
yield compared to the ligand-free reaction (61 vs. 59%;
Table 1, Entries 12 and 2), whereas DPPP [1,3-bis(diphenyl-
phosphanyl)propane] was the best ligand among those
screened, giving 80% yield (Table 1, Entry 9). In the ab-
sence of cobalt chloride only a trace amount of product was
[a] Isolated yield. [b] Reaction time was 4 h and ligand free.
[c] 30 mol-% catalyst was used and reaction time was 16 h.
[d] 15 mol-% catalyst was used. [e] Reaction time was 3 h. [f] Reac-
tion time was 20 h and reaction temperature was 140 °C. n.d. = not
determined.
In Table 3, the reactivity of various amines with 2-chloro-
obtained (Table 1, Entry 10). Attempts to lower the amount pyridines under cobalt catalysis is shown. The yield of the
of catalyst used were made, and the yield obtained was reaction between 2-chloropyridine and N-methylpiperazine
moderate for 5 mol-% catalyst, but for 1 mol-% catalyst the (50%; Table 3, Entry 1) was less than the one where piperi-
yield was only 5% (Table 1, Entries 13–15).
dine was employed (80%; Table 1, Entry 9), suggesting that
The reactions between piperidine and several N-aromatic extra binding of another nitrogen atom would not be bene-
chlorides are shown in Table 2. As it can be seen from this ficial for the reaction. Therefore, the reactivity of six-mem-
table, a nitrogen atom next to the carbon atom bearing a ber cyclic amines depended on the kind of heteroatom pres-
chloride group is essential for the reaction to take place ent, as shown in the case of N-methylpiperidine and morph-
(Table 2, Entries 1 and 5). In the case of 2-chloropyridines, oline. The yield was 83% in the case of morpholine
Eur. J. Org. Chem. 2009, 4586–4588
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
4587

G. Toma, K. Fujita, R. Yamaguchi
SHORT COMMUNICATION
Table 3. Reactions of 2-chloropyridine and 2-chloro-5-trifluoro- The mechanism of the reaction will be studied in the near
methylpyridine with miscellaneous amines in the presence of co-
future.
balt(II) chloride (10 mol-%) and DPPP (10 mol-%) at 135 °C in p-
xylene.
Experimental Section
General Procedure: To an oven-dried Schlenk tube that was flushed
with argon was consecutively added N-aromatic 2-chloride (1–
1.5 mmol), secondary amine (1–2 mmol), base (1 mmol), cobalt(II)
chloride (10 mol-%), phosphane ligand (10 mol-%), and p-xylene
(0.5 mL). The mixture was allowed to stir at 135 °C for 3 h in an
oil bath. After cooling the Schlenk tube to room temperature, cu-
mene was added as an internal standard and dichloromethane as a
solvent before GC analysis was performed. Isolation was done by
column chromatography on silica gel or alumina (hexane/ether).
Product identification was done by NMR spectroscopy, GC–MS,
and elemental analysis.
Supporting Information (see footnote on the first page of this arti-
cle): Full characterization data of the products.
[1] a) J. F. Hartwig, Acc. Chem. Res. 2008, 41, 1534–1544 and ref-
erences cited therein; b) H. Shinokubo, K. Oshima, Eur. J. Org.
Chem. 2004, 2081–2091.
[2] a) A. Correa, O. G. Mancheno, C. Bolm, Chem. Soc. Rev. 2008,
37, 1108–1117; b) B. D. Sherry, A. Furstner, Acc. Chem. Res.
2008, 41, 1500–1511; c) V. K. Gore, V. V. Ma, R. Tamir, N. R.
Gavva, J. J. S. Treanor, M. H. Norman, Bioorg. Med. Chem.
Lett. 2007, 17, 5825–5830.
[3] a) S. Narayan, T. Seelhammer, R. E. Gawley, Tetrahedron Lett.
2004, 45, 757–759; for reviews: P. Appukkuttan, E.
[a] Reaction time was 3 h. [b] GC yield using cumene as internal
standard. [c] Isolated yield. [d] Reaction time was 4 h. [e] 15 mol-%
cobalt chloride, ligand free. [f] 20 mol-% cobalt chloride and
20 mol-% DPPP were used.
Van der Eycken, Eur. J. Org. Chem. 2008, 1133–1155; b) A. Te-
wari, M. Hein, A. Zapf, M. Beller, Tetrahedron 2005, 61, 9705–
9709; M. S. Viciu, R. A. Kelly III, E. D. Stevens, F. Naud, M.
Studer, S. P. Nolan, Org. Lett. 2003, 5, 1479–1482; S. R.
Stauffer, S. Lee, J. P. Stambuli, S. I. Hauck, J. F. Hartwig, Org.
Lett. 2000, 2, 1423–1426; c) C. Desmarets, R. Schneider, Y.
Fort, J. Org. Chem. 2002, 67, 3029–3036; G. Manolikakes, A.
Gavryushin, P. Knochel, J. Org. Chem. 2008, 73, 1429–1434; d)
H. Zhang, Q. Cai, D. Ma, J. Org. Chem. 2005, 70, 5164–5173.
(Table 3, Entry 3), whereas in the case of N-methylpipera-
zine it was only 55% (Table 3, Entry 2). A fused piperidine
also reacted despite the slight steric hindrance to give good
yields (68 and 75%; Table 3, Entries 4 and 5). A noncyclic [4] a) E. Brenner, R. Schneider, Y. Fort, Tetrahedron 2002, 58,
6913–6924 and references cited there; b) C. Desmarets, R.
amine also reacted under the same conditions as shown in
Entry 6 to give the desired product in high yield (86%;
Table 3, Entry 6). Studies related to the reaction time were
Scheider, Y. Fort, J. Org. Chem. 2002, 67, 3029–3036.
[5] Y.-C. Teo, G.-L. Chua, Chem. Eur. J. 2009, 15, 3072–3075.
[6] a) D. Guo, H. Huang, J. Xu, H. Jiang, H. Liu, Org. Lett. 2008,
made, but further extension of the reaction time did not
give better results. Although the most relevant results are
shown here, other substrates were also tested and the scope
and limitations of the reaction is to be published in a future
full paper along with a more detailed investigation of the
reaction mechanism.
10, 4513–4516; b) A. Correa, C. Bolm, Angew. Chem. Int. Ed.
2007, 46, 8862–8865.
[7] a) H. Ohmiya, H. Yorimitsu, K. Oshima, J. Am. Chem. Soc.
2006, 128, 1886–1889 and references cited there; b) H. Ohmiya,
K. Wakabayashi, H. Yorimitsu, K. Oshima, Tetrahedron 2006,
62, 2207–2213.
[8] a) C. Gosmini, J.-M. Begouin, A. Moncomble, Chem. Com-
mun. 2008, 3221–3233 and references cited therein; b) M. Ama-
tore, C. Gosmini, Angew. Chem. Int. Ed. 2008, 47, 2089–2092.
[9] a) P. Shukla, Y.-C. Hsu, C.-H. Cheng, J. Org. Chem. 2006, 71,
655–658; b) B. Sezen, D. Sames, Org. Lett. 2003, 5, 3607–3610;
c) G. Dunet, P. Knochel, Synlett 2007, 9, 1383–1386 and refer-
ences cited there; d) G. Cahiez, C. Chaboche, C. Duplais, A.
Moyeux, Org. Lett. 2009, 11, 277–280; e) E. Shirakawa, T. Sato,
Y. Imazaki, T. Kimura, T. Hayashi, Chem. Commun. 2007,
4513–4515; f) M. Lombardo, S. Licciulli, F. Pasi, G. Angelici,
C. Trombini, Adv. Synth. Catal. 2005, 347, 2015–2018; g) A.
Kuno, N. Saino, T. Kamachi, S. Okamoto, Tetrahedron Lett.
2006, 47, 2591–2594 and references cited therein.
Conclusions
In summary, we have developed a new methodology to
prepare N-aromatic-containing tertiary amines that are use-
ful building blocks for bioactive molecules. Although sig-
nificant progress has been made during the past few years
in cobalt chemistry, the use of N-aromatic 2-chloro com-
pounds in C–N bond-forming reactions is reported for the
first time in this paper. Among the bases and ligands scre-
ened, potassium carbonate and DPPP gave the best results.
Received: June 1, 2009
Published Online: August 10, 2009
4588
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 4586–4588