A Convenient Procedure for the Oxidative Dehydrogenation of N-Heterocycles Catalyzed by FeCl2/DMSO

Article abstract of DOI:10.1055/s-0035-1561613

A convenient catalytic procedure has been developed for the oxidative dehydrogenations of N-heterocycles. Combining catalytic FeCl₂ with DMSO yields a catalyst that promotes the dehydrogenation of tetrahydroquinolines and related heterocycles under 1 bar of O₂, affording the corresponding N-heteroaromatic products in moderate yields.

Full text of DOI:10.1055/s-0035-1561613

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2016, 27, A–D
letter
A
W. Zhou et al.
Letter
Synlett
A Convenient Procedure for the Oxidative Dehydrogenation of
N-Heterocycles Catalyzed by FeCl2/DMSO
Weiyou Zhou
Piyada Taboonpong
Ahmed Hamdoon Aboo
Lingjuan Zhang
Jun Jiang
Jianliang Xiao*
Department of Chemistry, University of Liverpool,
Liverpool L69 7ZD, UK
jxiao@liv.ac.uk
Received: 15.02.2016
Accepted after revision: 18.03.2016
Published online: 20.04.2016
example of homogeneous dehydrogenation of tetrahydro-
quinolines using a [Cp*Ir(2-hydroxypyridine)] catalyst.15
More recently, Jones et al.16 reported an iron–pincer cata-
lyst that showed high catalytic activity in the dehydrogena-
tion of N-heterocycles; the same group also found a cobalt–
pincer catalyst for the same transformation.17 Although
these catalysts have advanced N-heterocycle dehydrogena-
tion considerably, their performance depends on the use of
specialist ligands. Thus, the development of inexpensive
and convenient catalytic systems for dehydrogenation is
still desirable. Herein we report a simple FeCl2/DMSO sys-
tem that catalyses efficient oxidative dehydrogenation of N-
heterocycles with molecular oxygen as the terminal oxi-
dant. To the best of our knowledge, there appears to be only
one report in which an iron salt without ligands,
Fe(NO3)3/TEMPO, was used for the oxidative dehydrogena-
tion of 1,2,3,4-tetrahydroquinoline, affording quinoline in a
yield of only 27%.18
We started the investigation choosing 1,2,3,4-tetrahy-
droquinoline as a model substrate. Preliminary screening of
iron salts showed that various iron compounds, for exam-
ple, FeCl2, FeBr2, and Fe(OTf)3 (3 mol%), could catalyse the
reaction without any ligands under 1 bar of O2 at 140 °C
(Table 1, entries 1–4). FeCl2 led to the highest product yield,
which is, however, only moderate. We then varied the sol-
vent and examined the reaction at a reduced temperature
(Table 1, entries 5–10). Interestingly, the substrate was fully
consumed when DMSO was used as the solvent, although
the yield of quinoline was not improved. The higher conver-
sion in DMSO could stem from the solvent acting as an oxi-
dant.19 Indeed, in the absence of the iron salt, quinoline was
obtained in 23% yield (Table 1, entry 11). These results
prompted us to reexamine the iron-catalysed dehydrogena-
tion in the presence of DMSO in p-xylene. Gratifyingly,
when the dehydrogenation was carried out at 110 °C with
DOI: 10.1055/s-0035-1561613; Art ID: st-2016-d0106-l
Abstract A convenient catalytic procedure has been developed for the
oxidative dehydrogenations of N-heterocycles. Combining catalytic FeCl2
with DMSO yields a catalyst that promotes the dehydrogenation of tet-
rahydroquinolines and related heterocycles under 1 bar of O2, affording
the corresponding N-heteroaromatic products in moderate yields.
Key words oxidative dehydrogenation, N-heterocycle, iron catalyst,
DMSO, quinolines
Catalytic dehydrogenation of saturated N-heterocycles
provides an easy, atom-economic entry to the correspond-
ing aromatic heterocycles. There are mainly two types of
dehydrogenation reaction. One is run under acceptorless
conditions, with the hydrogen released as H2. Catalysts such
as iridium–pincer complexes,1 Cu/Al2O3,2 Cu/TiO2,3 hy-
droxyapatite-bound palladium,4 and ruthenium–hydride
complexes5 have been explored for the acceptorless dehy-
drogenation. The other requires the presence of an oxidant
such as O2. This oxidative dehydrogenation has been per-
formed with a number of catalysts, including Ru/Al2O3,6
[Rh2(caprolactamate)4],7 Ru/Co3O4,8 Ru(OH)x/Al2O3,9 rhodi-
um nanoparticles supported on multiwalled carbon nano-
tubes,10 iron oxides surrounded by nitrogen-doped
graphene shells immobilized on carbon (FeOx@NGr–C),11
cobalt oxide supported on nitrogen-doped carbon,12 and
ο-quinone.13 It is noted that most of the catalytic systems
are based on noble metals or require harsh conditions, such
as high reaction temperatures, high pressure, and compli-
cated preparation methods.
We have recently reported highly efficient acceptorless
dehydrogenation of N-heterocycles with iridicycle cata-
lysts.14 Prior to this, Fujita and Yamaguchi reported the first
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D

B
W. Zhou et al.
Letter
Synlett
0.8 equivalents of DMSO added, quinoline was isolated in a
higher yield of 65% (Table 1, entry 19), indicating a syner-
gistic effect between FeCl2 and DMSO in the reaction.
tion of both electronic and steric effects (Table 2, entry 13).
However, installing a phenyl moiety at the 2 position con-
fers a much higher reactivity on the substrate, which was
fully reacted in 8 h (Table 2, entry 14). This is likely due to
the sp3 C–H bond being made weaker by the phenyl ring,
indicating that electronic effects may play a more import-
ant role than steric hindrance in the oxidative dehydroge-
nation reaction.
Table 1 Optimization of Conditions for the Dehydrogenation of
1,2,3,4-Tetrahydroquinolinea
O2
iron salt, DMSO
N
N
H
Table 2 Scope of the Oxidative Dehydrogenation of N-Heterocycles
under FeCl2/DMSOa
Entry
Iron salt
Solvent
DMSO
(equiv)
Temp
(°C)
Yield
(%)b
Entry
Substrate
Product20
Time
h)
Yield
(%)b
1
2
FeCl2
FeBr2
Fe(BF4)2
Fe(OTf)3
FeCl2
FeCl2
FeCl2
FeCl2
FeCl2
FeCl2
–
p-xylene
p-xylene
p-xylene
p-xylene
MeCN
0
140
140
140
140
100
100
100
100
100
100
100
100
100
100
100
100
80
45
26
8
0
3
0
1
2
3
24
24
24
70
65
82
N
H
N
4
0
23
25
12
5
0
6
EtOAc
0
N
H
N
N
7
THF
0
trace
8
DMF
0
13
42
24
23
41
48
52
46
13
32
42
65
9
DMSO
–
N
H
10
11
12
13
14
15
16
17
18
19
p-xylene
DMSO
0
–
FeCl2
FeCl2
FeCl2
FeCl2
FeCl2
FeCl2
FeCl2
FeCl2
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
p-xylene
3
4
5
24
24
75
54
N
H
1.5
0.8
0.3
0.15
0.8
0.8
0.8
N
N
H
N
N
90
6
7
24
24
73
59
110
N
H
a Reaction conditions: 1,2,3,4-tetrahydroquinoline (0.5 mmol), amount of
catalyst (3 mol%), 24 h, under 1 bar O2 atmosphere, solvent (1 mL).
b Isolated yields are given.
O
O
N
N
H
Using the optimized FeCl2/DMSO catalyst system (Table
1, entry 19), various derivatives of 1,2,3,4-tetrahydroquino-
line were dehydrogenated under a balloon pressure of O2.
As shown in Table 2, substituents at the 2, 3 or 4 positions
were well tolerated, with 2-methylquinoline, 3-meth-
ylquinoline, and 4-methylquinoline being produced in 65%,
82%, and 75% yields, respectively (Table 2, entries 2–4).
Quinolines with a substituent on the aromatic ring were
also obtained in comparable yields (Table 2, entries 7–13).
In addition, a 2,5,7-trisubstituted substrate was tested, and
the desired product was obtained in good yield (Table 2, en-
try 5). From the reaction time needed for different sub-
strates, it can be observed that electron-withdrawing
groups tend to deactivate the substrate (Table 2, entries 8–
12), although the effect varies and is not dramatic under the
conditions used. The 8-chloro-substituted tetrahydroquin-
oline necessitated a much longer time, presumably a reflec-
Cl
Cl
Br
F
8
9
28
26
24
24
71
71
52
76
N
N
H
Br
F
N
H
N
10
11
N
N
H
Cl
Cl
N
N
H
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D

C
W. Zhou et al.
Letter
Synlett
Table 2 (continued)
mized reaction conditions (Table 3). Although most of the
yields are lower than those in previous reports,6,7,12 some
substrates, such as acridine and quinoxaline, afforded the
target products in comparable yields with those obtained
with FeOx@NGr–C.11
Entry
Substrate
Product20
Time
h)
Yield
(%)b
Br
Br
12
30
40
81
64
To gain some insight into the mechanism of the
FeCl2/DMSO-catalyzed dehydrogenation of N-heterocycles,
the oxidation of 1,2,3,4-tetrahydroquinoline in the pres-
ence of 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) and
2,6-di-tert-butyl-4-methylphenol (BHT) were investigated.
Table 4 (entries 2 and 3) shows that TEMPO suppressed the
reaction to some extent, whilst BHT had little effect, indi-
cating that radicals may be involved at some stages of the
reaction. Control experiments showed that, without O2, the
reaction did not happen (Table 4, entries 4). This is in con-
trast with the iridicycle-catalysed dehydrogenation we re-
ported recently,14 where no oxidant is needed, with the hy-
drogen being released as H2. Thus, the reaction in question
is oxidative in nature, with the hydrogen being turned into
water. In a further experiment, N-methylquinoline was
used as the substrate. However, no dehydrogenation prod-
uct could be detected, indicating that the presence of the
N–H proton is critical for the dehydrogenation to proceed.12
N
N
H
13
14
N
H
N
Cl
Cl
8
80
N
H
N
a Reaction conditions: substrate (0.5 mmol), FeCl2 (3 mol%), DMSO (0.4
mmol), 110 °C, under O2 atmosphere, p-xylene (1 mL).
b Isolated yields are given.
The high reactivity of the tetrahydroquinoline deriva-
tives encouraged us to explore the FeCl2/DMSO catalytic
system for the dehydrogenation of other N-containing com-
pounds. As shown in Table 3, apart from quinoline deriva-
tives, quinoxaline, acridine, quinoxaline, isoquinoline, and
indole were also generated in 41–79% yield under the opti-
Table 4 Further Experiments for the Dehydrogenation of 1,2,3,4-Tet-
rahydroquinolinea
Table 3 Oxidative Dehydrogenation of Other N-Containing Com-
Entry
Catalyst
Additive
Conv. (%)
Yield (%)
pounds under FeCl2/DMSOa
1
2
3
4
FeCl2/DMSO
FeCl2/DMSO
FeCl2/DMSO
FeCl2/DMSOc
–
100
66
65
45
69
0
TEMPOb
BHTb
–
Entry
1
Substrate
Product
Time (h)
24
Yield (%)b
41
100
0
a Reaction conditions: 1,2,3,4-tetrahydroquinoline (0.5 mmol), FeCl2 (3
mol%), DMSO (0.4 mmol), reaction temperature 110 °C, 24 h, under O2 at-
mosphere, p-xylene (1 mL). Isolated yields are given.
b 1.0 mmol.
N
H
N
c Without O2.
2
3
20
24
79
47
N
H
N
According to the literature14 and our results, a possible
reaction pathway for the oxidative dehydrogenation of N-
heterocycles under FeCl2/DMSO is proposed and shown in
Scheme 1. Oxidative dehydrogenation of the tetrahydro-
quinoline, which may start from single electron transfer
from the nitrogen to the iron, affords an imine or iminium
intermediate in equilibrium with an enamine. The activated
imine is then reduced by the electron-rich enamine, fur-
nishing the starting substrate and quinoline.
In summary, we have developed a convenient protocol
for the oxidative dehydrogenation of N-heterocycles that
centres on the use of FeCl2 and substoichiometric DMSO.
Various substituted 1,2,3,4-tetrahydroquinolines and some
other N-containing compounds have been examined, giving
the desired products in moderate yields. Oxygen is indis-
pensible for the dehydrogenation and it appears that FeCl2
and DMSO act in concert to promote it.
NH
N
H
N
N
N
4
5
6
26
34
24
46
61
75
N
H
N
H
N
H
H
N
N
N
H
N
a Reaction conditions: substrate (0.5 mmol), FeCl2 (3 mol%), DMSO (0.4
mmol), 110 °C, under O2 atmosphere, p-xylene (1 mL).
b Isolated yields are given.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D

D
W. Zhou et al.
Letter
Synlett
(7) Choi, H.; Doyle, M. P. Chem. Commun. 2007, 745.
FeCl2
(8) Li, F.; Chen, J.; Zhang, Q. H.; Wang, Y. Green Chem. 2008, 10, 553.
(9) Kamata, K.; Kasai, J.; Yamaguchi, K.; Mizuno, N. Org. Lett. 2004,
6, 3577.
DMSO
N
N
O2
N
H
H
M
(10) Jawale, D. V.; Gravel, E.; Shah, N.; Dauvois, V.; Li, H. Y.;
Namboothiri, I. N. N.; Doris, E. Chem. Eur. J. 2015, 21, 7039.
(11) Cui, X. J.; Li, Y. H.; Bachmann, S.; Scalone, M.; Surkus, A.-E.;
Junge, K.; Topf, C.; Beller, M. J. Am. Chem. Soc. 2015, 137, 10652.
(12) Iosub, A. V.; Stahl, S. S. Org. Lett. 2015, 17, 4404.
(13) Wendlandt, A. E.; Stahl, S. S. J. Am. Chem. Soc. 2014, 136, 11910.
(14) (a) Wu, J. J.; Talwar, D.; Johnston, S.; Yan, M.; Xiao, J. L. Angew.
Chem. Int. Ed. 2013, 52, 6983. (b) Talwar, D.; Gonzalez-de-
Castro, A.; Li, H. Y.; Xiao, J. L. Angew. Chem. Int. Ed. 2015, 54,
5223.
N
Scheme 1 A possible pathway for the oxidative dehydrogenation of
tetrahydroquinolines catalyzed by FeCl2/DMSO where M may be Fe2+ or H+
Acknowledgment
(15) Yamaguchi, R.; Ikeda, C.; Takahashi, Y.; Fujita, K. J. Am. Chem.
Soc. 2009, 131, 8410.
(16) Chakraborty, S.; Brennessel, W. W.; Jones, W. D. J. Am. Chem.
Soc. 2014, 136, 8564.
We thank Changzhou University (WZ) and the Thailand Research
Fund (PT) for research fellowships.
(17) Xu, R. B.; Chakraborty, S.; Yuan, H. M.; Jones, W. D. ACS Catal.
2015, 5, 6350.
(18) Zhang, E. L.; Tian, H. W.; Xu, S. D.; Yu, X. C.; Xu, Q. Org. Lett.
2013, 15, 2704.
(19) MacGregor, W. S. Ann. N. Y. Acad. Sci. 1967, 141, 3.
(20) Typical Procedure
Supporting Information
Supporting information for this article is available online at
http://dx.doi.org/10.1055/s-0035-1561613.
S
u
p
p
ortioInfgrmoaitn
S
u
p
p
ortiInfogrmoaitn
To a Schlenk tube equipped with a magnetic stir bar were added
8-methyl-1,2,3,4-tetrohydroquinoline (0.50 mmol), FeCl2 (1.9
mg, 1.5·10–2 mmol), DMSO (31.2 mg, 0.4 mmol), and p-xylene (1
mL). The reaction mixture was stirred at 110 °C under an
oxygen atmosphere using a balloon and monitored by TLC. After
the reaction, the mixture was cooled to room temperature and
purified using flash chromatography (hexane–EtOAc, 10:1) to
give the corresponding product 8-methylquinoline in 70% yield.
8-Methylquinoline
References and Notes
(1) (a) Wang, Z. H.; Tonks, I.; Belli, J.; Jensen, C. M. J. Organomet.
Chem. 2009, 694, 2854. (b) Yao, W. B.; Zhang, Y. X.; Jia, X. Q.;
Huang, Z. Angew. Chem. Int. Ed. 2014, 53, 1390.
(2) Damodara, D.; Arundhathi, R.; Likhar, P. R. Adv. Synth. Catal.
2014, 356, 189.
(3) Mikami, Y.; Ebata, K.; Mitsudome, T.; Mizugaki, T.; Jitsukawa,
K.; Kaneda, K. Heterocycles 2011, 82, 1371.
(4) Hara, T.; Mori, K.; Mizugaki, T.; Ebitani, K.; Kaneda, K. Tetrahe-
dron Lett. 2003, 44, 6207.
(5) Muthaiah, S.; Hong, S. H. Adv. Synth. Catal. 2012, 354, 3045.
(6) (a) Yamaguchi, K.; Mizuno, N. Chem. Eur. J. 2003, 9, 4353.
(b) Yamaguchi, K.; Mizuno, N. Angew. Chem. Int. Ed. 2003, 42,
1480.
Colorless oil. 1H NMR (400 MHz, CDCl3): δ = 8.93 (m, 1 H), 8.10
(m, 1 H), 7.64 (d, J = 4.0 Hz, 1 H), 7.54 (m, 1 H), 7.43–7.35 (m, 2
H) 2.82 (s, 3 H). 13C NMR (100 MHz, CDCl3): δ = 149.2, 147.3,
137.1, 136.3, 129.6, 128.3, 126.3, 125.9, 120.8, 18.2. HRMS: m/z
calcd for [C10H9N + H+]: 144.0813; found: 144.0813.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2016, 27, A–D