Efficient 2,2,6,6-tetramethylpiperidine-1-oxyl/iron catalyzed aerobic dehydrogenation of hexanitrobibenzyl to Hexanitrostilbene

Article abstract of DOI:10.1002/jccs.201100473

Hexanitrostilbene (HNS) was efficiently produced through the dehydrogenation of hexanitrobibenzyl (HNBB) with oxygen catalyzed by 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)/ferrous chloride (FeCl₂) in 80% yield, and up to a 308 turnover number was achieved. First, TEMPO is used in the dehy-drogenation and the influence of reaction time, temperature, solvent and the catalyst were discussed. A possible mechanism of this catalytic process is proposed and it is obtained that Fe(II) can abstract hydrogen from HNBB and the function of the function of TEMPO is to oxidize Fe(II)OOH to Fe(III)OO..

Full text of DOI:10.1002/jccs.201100473

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
Article
Efficient 2,2,6,6-Tetramethylpiperidine-1-oxyl/iron Catalyzed Aerobic Dehydrogenation
of Hexanitrobibenzyl to Hexanitrostilbene
Ting-ting Lu and Ming Lu*
School of Chemical Engineer, Nanjing University of Science & Technology, Nanjing 210094, P. R. China
(Received: Aug. 2, 2011; Accepted: Mar. 1, 2012; Published Online: ??; DOI: 10.1002/jccs.201100473)
Hexanitrostilbene (HNS) was efficiently produced through the dehydrogenation of hexanitrobibenzyl
(
(
HNBB) with oxygen catalyzed by 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO)/ferrous chloride
FeCl ) in 80% yield, and up to a 308 turnover number was achieved. First, TEMPO is used in the dehy-
2
drogenation and the influence of reaction time, temperature, solvent and the catalyst were discussed. A
possible mechanism of this catalytic process is proposed and it is obtained that Fe(II) can abstract hydro-
.
gen from HNBB and the function of the function of TEMPO is to oxidize Fe(II)OOH to Fe(III)OO .
2
Keywords: Aerobic dehydrogenation; TEMPO; Fe ; HNS.
+
INTRODUCTION
lytical pure grade. All the reagents were used without fur-
ther purification. NMR spectra were recorded on a Bruker
500-MHz spectrometer using dimethyl sulfoxide (DMSO)
as the solvent with tetramethylsilane (TMS) as an internal
standard. IR spectra were recorded on a Bruker Vector 22
infrared spectrometer, as KBr pellets with absorption in
2
,2¢,4,4¢,6,6¢-Hexanitrostilbene (HNS) is a thermally
stable explosive which is extensively used in petroleum
1
production and military ordnance systems. The main
4
methods to prepare HNS are the dehydrogenation of its in-
-3
termediate hexanitrobienzyl (HNBB) by halogenating
agents, quinones, or other stoichiometric reagents. How-
ever, these methods suffer from the inherent problem of us-
ing carcinogenic oxidant and severe environmental pollu-
tion. Molecular oxygen was attempted as oxidant in the de-
-
1
cm . High performance liquid chromatography (HPLC)
experiments were performed on a liquid chromatograph
(Dionex Softron GmbH, America), consisting of a pump
(P680) and ultraviolet–visible light detector (UVD) system
(170U). Diacovery C18 column was used for the HPLC
analysis with amobile phase consisting of acetonitrile and
4
hydrogenation, but it failed.
The stable tetraalkylnitroxyl radical 2,2,6,6-tetra-
methylpiperidine-1-oxyl (TEMPO) which is a well-known
oxidation catalyst has numerous applications in organic
synthesis. It can combine with either transitional metals
1
5
methanol (5:5, v/v).
Experimental procedure
1
HNBB was prepared according to the literature. IR
6
5
-6
7
8
copper, iron, and ruthenium ) or nonmetal catalyst
9-10
(
(
(KBr): u 3103, 2888, 1821, 1641, 1545, 1468, 1356, 1226,
1
1
12
-1 1
nitrite ) to catalyze the aerobic oxidation of alcohols,
1
1166, 1089, 916, 804, 710 cm . H NMR (DMSO-d
3.344 (s, 4H, CH ), 9.098 (s, 4H, Ar-H).
A mixture of HNBB (2.21*10 mol), TEMPO (1.3
6
): d
3
14
polysaccharide, and organometallic compounds. Here-
in, we report a novel approach for the aerobic dehydrogen-
2
-
3
-
4
-4
ation of HNBB to HNS with TEMPO/FeCl
2
catalytic sys-
*10 mol), and FeCl
were added in a 100 mL four-neck round flask equipped
with magnetic stirring apparatus and water-bath. O was
2
(1.44*10 mol) in DMSO (10 mL)
tem in the yield of 80%, and the reaction mechanism is pro-
posed. To the best of our knowledge, no literature has been
reported that TEMPO is used to catalyze the dehydrogena-
tion.
2
bubbled into the flask at the flow rate of 5 mL/h. The reac-
tion mixture was stirred for 8 h at 55 °C and poured into ex-
cess water. After standing for about 2 h to allow formation
of the precipitate, the solid was filtered off. Then it was
washed with hot acetone and dried at 55 °C to give pale yel-
low HNS (0.81 g, yield 80%). IR (KBr): u 3102, 2894,
1610, 1600, 1534, 1477, 1341, 1268, 1176, 1075, 960, 917,
EXPERIMENTAL
Materials and methods
TEMPO was purchased from Acros Organics with
chemically pure. Other reagents were purchased from
Sinopharm chemical reagent, Shanghai (China) with ana-
-1 1
794, 722, 556, 514 cm . H NMR (DMSO-d
6
): d 7.142 (s,
*
Corresponding author. E-mail: tt252525@gmail.com,luming302@126.com
J. Chin. Chem. Soc. 2012, 59, 000-000
© 2012 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Article
Lu and Lu
2
H, CH), 9.111 (s, 4H, Ar-H).
Table 1. The influence of different solvents on dehydrogenation
of HNBB
RESULTS AND DISCUSSION
The volume The yield of
Run
Solvents
(mL)
HNS (%)
Influences of the reaction temperature and time on
the dehydrogenation
1
2
3
4
5
6
7
8
9
ethyl acetate
benzene/ethanol
acetonitrile
15
/
15/15
20
/
The influences of reaction temperature and time are
shown in Fig. 1. As illustrated, with the increase in temper-
ature the yield of HNS increased and reached a maximum
at 55 °C, but then decreased when the temperature further
17.92
acetone
15
12.25
DMSO
15
68.70
Tetramethylene sulfone
DMF
15
/
1
increased. Sun had mentioned that nitro group had the
7
15
/
/
cyclohexylamine
Hexamethylphosphoramide
15
ability to act as radical scavenger and quench the radical re-
15
Trace
1
action. Similarly, Yasushi had observed a negative influ-
8
ence of nitro groups on their reaction system when sub-
strates with NO
2
were used. As the dehydrogenation was
amine (cyclohexylamine) as the solvent brought to lots of
unknown side products (Run 7-9). On the basis of these re-
sults, DMSO was thought to be the best solvent for the
dehydrogenation. A likely explanation was that DMSO can
not only dissolve the substrate and catalyzer, but also bond
radical reaction and TEMPO did not lose the activity in
9
,11
high temperature, a possible explanation was that the ni-
tro group in HNBB may quench the reaction when the tem-
perature was higher.
1
9-20
From Fig. 1, it can be seen the yield was increased as
the reaction time was longer, and it reached 80.03% at 8 h.
Further elongating the reaction, there was no influence on
the yield. This phenomenon indicated that the reaction
fully completed within 8 h.
to metal ion (such as iron) through sulfur and oxygen
which increased the catalyst activity of the metal catalyst
Run 5). To confirm this point, tetramethylene sulfone
(
which has similar structure to DMSO but can not bond to
metal ion was used instead, but no product appeared (Run
Influence of the solvent on the dehydrogenation
Then a series of solvents were examined to choose the
suitable one (Table 1). Due to the poor solubility of HNBB
in ethyl acetate and benzene/ethanol, no products appeared
when they were employed (Run 1 and 2). Acetonitrile,
which is a good solvent in the TEMPO-catalyzed oxidation
6
).
The influence of the volume of DMSO was also stud-
ied (Fig. 2). When DMSO was less than 3 mL, HNBB can’t
be dissolved completely. With the amount of DMSO in-
creased, the yield increased and reached maximum at 10
mL. However, further addition of DMSO resulted in the de-
crease in the yield. The crude products were analyzed from
HPLC, and it revealed that the proportion of HNBB was
7
of alcohol, was not suit for the present oxidation (Run 3).
The boiling point of acetone was too low to be chosen (Run
4
). Using acylamide (DMF, hexamethylphosphoramide) or
Fig. 1. Influences of reaction time and temperature on
the dehydrogenation.
Fig. 2. Influence of the volume of DMSO on the de-
hydrogenation.
2
www.jccs.wiley-vch.de
© 2012 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Chin. Chem. Soc. 2012, 59, 000-000

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
Dehydrogenation of Hexanitrobibenzyl to Hexanitrostilbene
only 0.01% when DMSO was 10 mL but was nearly 40%
while DMSO was 25 mL. The reason was that too much
solvent may reduce the chance the catalyst contacted with
the reactant.
Table 2. Different metal salts catalyzed aerobic oxidation
a
dehydrogenation of HNBB
The yield of
HNS with
The yield of
HNS without
TEMPO (%)
Run Transition metal salts
TEMPO (%)
Influence of catalyst on the dehydrogenation
The choice of metal salts
1
K CO
2
11.95
24.85
24.80
22.90
37.83
39.82
80.98
79.35
35.89
39.82
38.82
32.86
62.72
43.65
34.85
41.82
39.88
34.85
7.88
11.99
11.01
/
3
.
2
MgSO H O
2
4
In order to optimize the yield of HNS, several metal
salts co-catalysts were screened in combination with
TEMPO (Table 2). In contrast to the relative catalytic activ-
ity of metal salts, most transition metal salts is higher than
s-block metal salts (Run 1 to 17). The most effective metal
3
BaCl2
.
4
Pb(CH COO) 3H O
3
20.99
/
2
2
.
3
5
CrCl
4H O
2
.
6
MnSO H O
2
20.99
20.56
21.22
trance
/
4
7
FeCl
.
4
2
8
FeSO 4H O
2
salt was FeCl
job (62.20%, Run 13). However, there was no product
when their higher valence metal salts FeCl or CuCl ×H
was employed (Run 19 and 20). The influence of anions
was also discussed. And the yield of HNS used FeSO was
similar to that of FeCl (Run 7 and 8). However, when halo-
2
(80.98%, Run 7) while CuCl also did good
.
9
CoCl 4H O
2 2
10
Co(acac) *
2
.
1
1
1
1
1
2
3
4
Co(CH COO) 4H O
2
/
O
3
2
3
2
2
.
NiCl 6H O
2
/
2
CuCl
34.99
33.99
30.99
13.01
11.99
34.99
8.99
/
4
CuBr
CuI
2
15
1
1
1
1
6
7
8
9
ZnCl
.
gen ions were employed, the yields decreased with the ra-
dius of halogen increased (Run 13 to 15).
2
ZnSO 7H O
4
2
.
3CdSO 8H O
2
4
Influence of the proportion of TEMPO in the total cat-
alyst
FeCl
.
3
20
CuCl H O
2
trace
2
The effect of the proportion of TEMPO is obtained in
Fig. 3. The yield of HNS was more than 20% in the absence
of TEMPO and it increased with enhancing the proportion
and reached a maximum at 40%. However, with the further
increase of the proportion, the yield had a slightly declined
because higher concentrations of TEMPO may inhibit the
a
-3
Metal salts were1.44*10 mol and other conditions were
immovable.
*
: Co(acac) is the abbreviation of cobalt(II) acetylacetonate.
2
than 10% only TEMPO added.
Influence of the amount of catalyst and the calcula-
tion of TON
2
1
dehydrogenation. It is strange that a small peak in the
yield appeared when the proportion reached 70%. When
the proportion was higher than 70%, the yield continued
The influence of the amount of FeCl
2
and TEMPO on
the dehydrogenation was obtained in Table 3 and Fig. 4, re-
2
decreasing due to the lack of FeCl , and the yield was less
Fig. 4. Influence of the amount of FeCl
2
on the de-
hydrogenation (1): The amount of FeCl was
.144 mmol and the reaction temperature was
5 °C.
Fig. 3. The proportion of TEMPO in the total cata-
(1)
lyst (1): the proportion of TEMPO = mol
(1)
2
0
5
(
2
TEMPO)/mol (FeCl +TEMPO).
J. Chin. Chem. Soc. 2012, 59, 000-000
© 2012 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jccs.wiley-vch.de
3

Article
Lu and Lu
Table 3. The influence of the amount of FeCl on the dehydro-
2
Table 4. The stoichiometric oxidation of HNBB*
a
genation
Run
TEMPO (mol)
FeCl (mol) The yield of HNS (%)
2
The amount of
Yield
TON*
Yield
TON
-3
Entry
1
2
3
4
2.21*10
/
10.04
/
FeCl (mol)
2
(80 °C) (80 °C) (55 °C) (55 °C)
-3
-3
/
2.21*10
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
6
6
6
5
5
5
5
5
5
5
5
5
5
5
4
4
4
4
-3
1
2
3
4
5
6
7
8
9
2.52*10
5.40*10
9.15*10
1.08*10
1.44*10
1.70*10
2.16*10
2.63*10
2.91*10
3.20*10
3.63*10
5.39*10
7.19*10
8.99*10
1.08*10
1.26*10
1.44*10
1.80*10
7.03
27.12
22.10
2.01
/
45.20
30.07
17.07
23.10
13.06
36.11
39.17
35.15
46.20
30.13
32.14
38.16
45.17
50.22
56.25
66.54
80.35
46.20
308.77
252.60
17.08
26.81
4.70
2.21*10
10.04
10.04
-3
70.08
29.23
/
2.21*10
2.21*10
*
: The reactions were in an inert (N ) atmosphere.
2
8.04
/
33.15
22.10
35.24
28.12
12.05
17.03
29.16
33.15
35.13
39.45
42.61
44.80
32.65
30.09
11.36
21.21
13.76
1.41
4.28
7.86
7.12
6.18
6.02
5.72
5.44
2.77
-4
33.91
29.84
21.13
27.49
1.76
2
2
0% as only catalysis amount FeCl and oxygen are both
employed.
HNBB into HNS is fulfilled by the coupling of dip-
.
loid redox reactions and Fe(III)OO is an active oxidant in
1
1
1
1
1
1
1
1
1
a
0
1
2
3
4
5
6
7
8
13.48
11.55
10.81
9.89
this system. The running of Cycle 1 continuously provides
.
the Fe(III)OO by oxidizing Fe(II)OOH with TEMPO,
24
2
5
and the latter is reduced into TEMPOH. One proof is that
HNS didn’t appear as the stoichiometric FeCl was used,
but it is nearly 20% when the oxygen and catalysis amount
of FeCl were employed. The oxidation of TEMPOH into
TEMPO is a process accomplished with molecular oxy-
9.25
2
9.92
10.93
5.64
2
The amount of TEMPO was 1.3*10 mol.
2
5
*
TON (turnover number) = (mol (HNS with FeCl /TEMPO)-mol
2
gen (Cycle 2), and the coupling of the two cycles fur-
nishes a novel and efficient aerobic dehydrogenation sys-
tem. TEMPO also can directly oxidize HNBB to HNS, but
the yield is poor (10%).
(
HNS with only TEMPO))/mol (FeCl2).
spectively. The yield of HNS was not regular either in 80
-
5
To deepen the research, the effects of only using
3+
°
C or 55 °C when FeCl
As FeCl reached 3.2*10 mol, the yield was higher with
the increase of FeCl
Further addition of FeCl
clined significantly the yield because high concentrations
2
was less than 3*10 mol (Table 3).
+
metal salts as the catalyst were studied (Table 3). K , Fe
-
5
2
2
and Cu were too steady to be oxidized to the correspond-
+
-
4
2
and arrived at maximum at 1.44*10 .
ing higher valence cations by oxygen so Equation 1 could
2
not happen. Pb can abstract H from HNBB, but it can’t be
2+
oxidized by TEMPO. Co and Ni were hardly directly
2
, under the same conditions, de-
+
2
+
2
2
of metal may act as an inhibitor for the reaction. The TON
was calculated and the highest value was 70.28 and 308.77
at 80 °C and 55 °C, respectively. On the contrary, the reac-
tivity of dehydrogenation increased with the increase in
oxidized by O and TEMPO accelerated the reaction rate
2
and increased the yield.
Scheme I Proposed mechanism for the aerobic de-
-
5
TEMPO amount until 1.3*10 mol, and then it decreased
with further addition.
hydrogenation of HNBB
The mechanism of the dehydrogenation
A lot of researchers referred the catalytic mechanism
1
0,11,23
of TEMPO in their articles.
In order to clarify the
mechanism of the Fe/TEMPO system, following experi-
ments were carried out (Table 4). First, HNBB was oxi-
dized by a stoichiometric amount of TEMPO in an inert
(
N
2
) atmosphere and the yield of HNS was only 10.04% no
matter how much FeCl was added (Run 1, 3 and 4). Then
the dehydrogenation was carried out in an inert (N
sphere with only a stoichiometric amount of FeCl
CONCLUSIONS
We have first examined the aerobic dehydrogenation
of HNBB to obtain HNS by the use of combined catalytic
2
2
) atmo-
and no
system of TEMPO with the transition metal salts like FeCl
2
2
HNS appeared (Run 2). It is indicated that oxygen is neces-
sary in the dehydrogenation because the yield is nearly
in 80% excellent yield. It was first reported TEMPO was
used in the dehydrogenation. The optimal reaction temper-
4
www.jccs.wiley-vch.de
© 2012 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Chin. Chem. Soc. 2012, 59, 000-000

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
Dehydrogenation of Hexanitrobibenzyl to Hexanitrostilbene
ature and time were 55 °C and 8 h, respectively; and DMSO
10. HerrerR, C. I. Tetrahedron Lett. 2006, 47, 13.
1. Miao, C. X.; Wang, J.; Wu, F. J. Org. Chem. 2010, 75, 257.
12. Inokuchi, T.; Torii, S. Tetrahedron Lett. 1995, 36, 3223.
1
2
was the most appropriate solvent. The amount of FeCl and
-
4
-4
TEMPO were 1.44*10 mol and 1.3*10 mol. A possible
mechanism was proposed and the function of Fe(II) was to
abstract hydrogen from HNBB and TEMPO can change
1
1
1
1
1
3. Isogai, T.; Isogai, A. Biomacromolecules 2010, 11, 1593.
4. Wei, W. Synth. Commun. 2010, 40, 1106.
5. Li, P. Chromatography 1993, 11, 180.
×
Fe(II)OOH to Fe(III)OO .
6. Gilbert, E. E. Propellants, Explos., Pyrotech. 1980, 5, 15.
7. Sun, Y.; Hu, X.; Li, H. J. Phys. Org. Chem. B. 2010, 114,
REFERENCES
4
862.
8. Yoshino, Y.; Iwahama, T.; Sakaguchi, S. J. Org. Chem. 1997,
2, 6810.
1
. Munir, I.; Hu, S.; Dordick, J. S. Adv. Synth. Catal. 2002, 344,
097.
1
1
6
2
. Williams, D. L.; Kuklenz, K. D. Propellants, Explos.,
Pyrotech. 2009, 34, 452.
1
2
2
9. Ahmed, M. A. J. Mater. Sci. Lett. 1989, 8, 104.
0. Dijksman, A. Org .Biomol. Chem. 2003, 1, 3232.
1. Laugier, J.; Caneschi, A.; Rey, P. Inorg. Chem. 1991, 30,
3
4
5
. Golding, P. US Patent 5,023,386, 1991.
. Gilbert, E. E. Propellants, Explos., Pyrotech. 1980, 5, 168.
. Michel, C.; Gamez, P.; Baerends, E. J. Inorg. Chem. 2009,
4
474.
2
2
2. Shilov, A. E.; Shulpin, G. B. Chem. Rev. 1997, 97, 456.
3. Baron, P.; Zhang, R.; Feigin, E.; Chevion, M. J. Am. Chem.
Soc. 1999, 121, 8070.
4
8, 11909.
6
7
. Jiang, N. J. Org. Chem. 2006, 71, 7087.
. Wang, N.; Chen, J.; Liang, X. Chem. Commun. 2005, 3,
2
2
4. Punniyamurthy, T.; Velusamy, S.; Iqbal, J. Chem. Rev. 2005,
5
322.
1
05, 2329.
8
9
. Dijksman, A. Chem. Commun. 1999, 1591.
5. Sheldon, R. A.; Adends, I. W. Acc. Chem. Res. 2002, 35, 774
. Liu, R.; Dong, C.; Hu, X. J. Am. Chem. Soc. 2004, 126, 4112.
J. Chin. Chem. Soc. 2012, 59, 000-000
© 2012 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jccs.wiley-vch.de
5