Model molecules with oxygenated groups catalyze the reduction of nitrobenzene: Insight into carbocatalysis

Article abstract of DOI:10.1002/cctc.201402070

The role of different oxygen functional groups on a carbon catalyst was studied in the reduction of nitrobenzene by using a series of model molecules. The carbonyl and hydroxyl groups played important roles, which may be ascribed to their ability to activate hydrazine. In comparison, the ester, ether, and lactone groups seemed to be inactive, whereas the carboxylic group had a negative effect. The reaction occurred most likely through a direct route, during which nitrosobenzene may be converted directly into aniline. Modeling carbon: Thanks to model molecules, the carbon-catalyzed reduction of nitrobenzene is mimicked. The role of different oxygen functional groups on a carbon catalyst is studied, and the carbonyl and hydroxyl groups seem to be the most important moieties, which may be ascribed to their ability to activate hydrazine. The reaction occurs more likely through a direct route, during which nitrosobenzene may also be converted directly into aniline.

Full text of DOI:10.1002/cctc.201402070

CHEMCATCHEM
COMMUNICATIONS
DOI: 10.1002/cctc.201402070
Model Molecules with Oxygenated Groups Catalyze the
Reduction of Nitrobenzene: Insight into Carbocatalysis
Shuchang Wu,[a, b] Guodong Wen,[a] Xiumei Liu,[c] Bingwei Zhong,[a] and Dang Sheng Su*[a]
The role of different oxygen functional groups on a carbon cat-
alyst was studied in the reduction of nitrobenzene by using
a series of model molecules. The carbonyl and hydroxyl groups
played important roles, which may be ascribed to their ability
to activate hydrazine. In comparison, the ester, ether, and lac-
tone groups seemed to be inactive, whereas the carboxylic
group had a negative effect. The reaction occurred most likely
through a direct route, during which nitrosobenzene may be
converted directly into aniline.
the carbon catalysts in the reactions and especially the actual
identities of the active sites are still under discussion. Mecha-
nistic research over carbon catalysts remains a challenge be-
cause of their complex surface structure and the co-existence
of various kinds of functional groups. In addition, undeter-
mined impurities may also contribute to the catalytic activity
of carbon materials.[12] Consequently, the application of
a model catalyst could be an effective method.[13]
Carbon materials such as graphite oxide, graphene, carbon
nanotubes, and active carbon can be considered as carbon
atoms grouped into layers of fused aromatic rings, and there
are commonly many kinds of oxygenated groups on their sur-
face. To study the role of various oxygenated groups in the se-
lective reduction of nitrobenzene, we selected a series of
For a long time, carbon was mainly used as a support for
metal catalysts,[1] but it is in fact also a catalyst in its own
right.[2] Recent years have witnessed the rapid development of
carbon-catalyzed reactions. Besides gas-phase reac-
tions, such as the oxidative dehydrogenation of eth-
ylbenzene and light alkanes,[3] carbon materials have
shown potential applications in several important
liquid-phase reactions that were conventionally cata-
lyzed by metal or metal oxide.[3] Bielawski and co-
workers found that graphite oxide was an efficient
catalyst for some oxidation and hydration reactions.[4]
The Kakimoto group developed a nitric acid assisted
nanoshell carbon-catalyzed method for the oxidation
of alcohols.[5] In addition, carbon materials also
played an important role in reactions such as the
ring opening of epoxides,[6] oxidative coupling,[7] and
oxidation of cyclohexane.[8]
Selective reduction of nitroarenes represents an
important method for the preparation of the corre-
sponding aromatic amines from both an industrial
and a scientific point of view. It was reported that
natural graphite,[9] fullerene,[10] and reduced graphene
oxide[11] could effectively catalyze the reduction of ni-
trobenzene. However, the detailed catalytic roles of
Scheme 1. Structures of different model catalysts.
model catalysts with different functional groups to mimic car-
bocatalysis (Scheme 1).
[a] S. Wu, Dr. G. Wen, B. Zhong, Prof. Dr. D. S. Su
Shenyang National Laboratory for Materials Science
Institute of Metal Research, Chinese Academy of Sciences
72 Wenhua Road, Shenyang 110016 (P.R. China)
E-mail: dssu@imr.ac.cn
The performance of each model catalyst is summarized in
Table 1. Phenanthraquinone exhibited the best performance as
far as nitrobenzene conversion was concerned. However, poor
selectivity for aniline was observed, and there was no clear im-
provement in the selectivity with a prolonged reaction time
(Table 1, entries 1 and 2). Surprisingly, a large amount of hy-
droxylamine was formed (ꢀ60%). The other byproducts were
mainly azoxybenzene and azobenzene (not shown). If the reac-
tion was catalyzed by 9,10-anthraquinone, the aniline selectivi-
ty increased sharply to over 98%, although with relatively mild
conversion (Table 1, entries 3 and 4). The results of these two
model catalysts indicated that the oxygen functional groups
[b] S. Wu
Graduate University of Chinese Academy of Sciences
19 Yuquan Road, Beijing 100049 (P.R. China)
[c] Dr. X. Liu
State Key Laboratory of Catalysis
Dalian Institute of Chemical Physics
Chinese Academy of Sciences
457 Zhongshan Road, Dalian 116023 (P.R. China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201402070.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 4
&
1
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
Table 1. Performance of the model catalysts in the reduction of nitro-
benzene.[a]
Entry
Catalyst
Conv.
[%]
Sel.
[%]
Yield
[%]
1[b]
2
A
A
B
B
C
D
E
F
G
H
I
J
–
99.0
100
22.0
21.8
22.1
81.4
96.1
72.3
55.3
54.2
61.0
25.2
22.0
21.0
15.6
24.8
95.3
22.1
98.5
98.4
94.2
98.7
99.1
80.2
99.9
99.9
3[b]
4
82.7
97.7
76.7
56.0
54.7
76.1
25.2
22.0
21.0
15.6
24.8
96.1
5
6
7
8
9
10
11
12
13[c]
14[d]
100
100
100
Figure 1. Plot of the catalytic performance against the loading of 9,10-an-
thraquinone. Reaction conditions: Nitrobenzene (2.4 g), hydrazine monohy-
drate (6.0 equiv.), ethanol (4.0 mL), 1008C, 20 min.
B
99.2
[a] General conditions: Functional groups (0.3 mmol), nitrobenzene
(1.2 g), hydrazine monohydrate (6.0 equiv.), ethanol (2.0 mL), 1008C, 5 h.
[b] 3 h. Unless otherwise noted, the reaction temperature was the tem-
perature of the oil bath. [c] No catalyst was used. [d] Under a helium
atmosphere.
Aiming to gain more insight into these model catalysts mim-
icking carbocatalysis, we then mainly studied 9,10-anthraqui-
none-catalyzed reactions owing to the excellent performance
of this catalyst. A relatively good linear correlation between
the catalyst loading and the initial catalytic activity was ob-
served (Figure 1); this indicated again that the carbonyl group
played an important role and that it was one type of active
species.
on anthraquinone-type catalysts were more suitable for the
aimed reaction under the selected conditions. However, the re-
sults also suggested that not only the identity of the oxygenat-
ed groups but also their chemical environments were critical
to the reaction, including the product distribution. Upon using
anthrone as the catalyst, which has only one carbonyl group,
both the conversion and aniline selectivity were lower than
that obtained with the use of 9,10-anthraquinone (Table 1,
entry 5). As the same amount of functional groups were added
in the reactions for the two catalysts, these results revealed
that conjugated carbonyl groups were more effective than two
isolated ones.
We assumed that hydrazine was first activated by the oxy-
genated groups. Taking the carbonyl group as an example, the
oxygen atom that has unpaired electrons can interact with the
hydrogen atom of a hydrazine molecule. As a result, the NÀH
bond is weakened, which is beneficial to its decomposition. It
was found that the activity improved notably if the reaction
was performed in DMSO (Table S1, entries 12 and 13), possibly
because the S=O group in DMSO interacted well with the hy-
drogen atoms of hydrazine thus to weaken the NÀH bonds.
The activity was further improved upon performing the reac-
tion in [Bmim]Cl (Bmim=1-butyl-3-methylimidazolium), which
was reported to have good ability to dissolve cellulose be-
cause the chloride ions can interact with the hydrogen atoms
in the cellulose molecules (Table S1, entries 14 and 15).[14] This
result confirmed the importance of the formation of hydrogen
bonds in the activation of hydrazine. However, as the hydra-
zine decomposition tests indicated, the active sites may only
activate hydrazine. Hydrazine started to decompose if both the
catalyst and nitrobenzene existed in the mixture (Figure S5).
For reduction of C=C bonds by using hydrazine as the re-
ducing agent, the actual reducing agent is diimide, which is
generated by oxidation of hydrazine.[15] However, for the re-
duction of the nitro group in the present study, the reaction
proceeded under an atmosphere of helium as well as in air
(Table 1, entry 14), which suggests that the active hydrogen re-
sulting from hydrazine decomposition was the real reducing
species and not diimide.
1,4-Benzoquinone and 1,4-benzenediol exhibited almost the
same conversion as well as the same aniline selectivity
(Table 1, entries 6 and 7). Analysis by NMR spectroscopy
showed that after hydrazine was added, 1,4-benzoquinone
gave the same 1H NMR and 13C NMR spectra as 1,4-benzenediol
(Figure S1, Supporting Information). Given that 1,4-benzenediol
is a weak acid and that hydrazine served as the base, the
former could be mainly dissociated and converted into
[C6H4O2]2À. In addition, UV/Vis spectroscopy (Figure S2) also
confirmed this transformation. 9-Phenanthrenol was used to
study the role of the hydroxyl group further, and the results
showed that the hydroxyl group was indeed active for this re-
action (Table 1, entry 8).
Benzyl benzoate, benzyl ether, phthalide, and terephthalic
acid were chosen to mimic the ester, ether, lactone, and car-
boxylic acid groups, respectively (Table 1. entries 9–12). Benzyl
benzoate, benzyl ether, and phthalide were not active relative
to the blank test within experimental error, but terephthalic
acid exhibited the lowest activity, and no improvement was
achieved at higher catalyst loading (Figure S4). The low activity
of the carboxylic group may be ascribed to the fact that this
hydrophilic group hinders the adsorption of nitrobenzene.
We further studied the reaction routes in this metal-free cat-
alyzed reduction process. The reaction routes included direct
and condensation pathways, which are generally accepted for
the reduction of nitroarenes.[16] For the direct pathway, nitro-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 4
&
2
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
benzene is reduced to nitrosobenzene, hydroxylamine, and
aniline, successively, whereas for the condensation pathway, ni-
trosobenzene reacts with hydroxylamine to form azoxyben-
zene, which is then reduced to azobenzene, hydrazobenzene,
and aniline, successively.
A plot of the reaction products as a function of reaction
time is depicted in Figure 2, and it shows that the aniline selec-
tivity was maintained at approximately 98% over the whole
Figure 3. Evolution of azoxybenzene and azobenzene. Reaction conditions:
9,10-Anthraquinone (0.15 mmol), substrate (0.3 g), hydrazine monohydrate
(12.0 equiv.), ethanol (2.0 mL), 1008C. C is the concentration of the reactant,
whereas C0 is the initial concentration of the reactant (five experiments were
performed for each substrate).
even at 1008C for 30 min (Table S2, entry 5); this indicated that
hydroxylamine was converted much more slowly than nitroso-
benzene. This phenomenon suggested that nitrosobenzene
was converted very fast, nitrosobenzene may be reduced di-
rectly to aniline without the formation of intermediate hydrox-
ylamine, and the conversion of nitrosobenzene was a noncata-
lytic process. On the basis of these findings, we proposed
a possible reaction pathway, as shown in Scheme 2.
Figure 2. Evolution of nitrobenzene and aniline. Reaction conditions: 9,10-
Anthraquinone (0.15 mmol), nitrobenzene (1.2 g), hydrazine monohydrate
(6.0 equiv.), ethanol (2.0 mL), 1008C (six experiments were performed for re-
actions at 1, 2, 3, 4, 5, and 6 h).
process. Even at the beginning stage, the selectivity of aniline
was already above 90%, which is indicative of the quick forma-
tion of aniline. In addition, first-order reaction kinetics were ob-
served for the overall reaction (Figure S6).
Although there was only a small portion of azoxybenzene
and related products, we still could not exclude the possibility
that azoxy-, azo-, and hydrazobenzene served as intermediates,
because they may be converted very fast into aniline. To clarify
this, we used these suspected intermediates as the initial reac-
tants (Figure 3). It was shown that even after 5 h, there was
still a large portion of azoxybenzene (27.9%) and azobenzene
(66.9%) that remained. The dominant product was hydrazo-
benzene, and no aniline was detected. Then, we used hydrazo-
benzene as a substrate and found that its concentration was
slightly increased, which probably resulted from the reduction
of azobenzene, because the reagent we obtained contained
approximately 10% azobenzene (Table S2, entry 1). The afore-
mentioned results indicated that the three suspected inter-
mediates could not be converted into aniline; thus, the con-
densation route could be neglected.
Scheme 2. Proposed reaction route for the reduction of nitrobenzene cata-
lyzed by 9,10-anthraquinone (r3@r1, r3@r2).
Notably, nitrobenzene may also be converted directly into
hydroxylamine under certain conditions considering that
a large amount of hydroxylamine was formed if phenanthra-
quinone was used as the catalyst. As a matter of fact, nitroben-
zene could be reduced directly to hydroxylamine by using
a metal catalyst.[16,17]
On the basis of these findings, we thus proposed a simple
mechanism for the reduction of nitrobenzene catalyzed by
carbon model catalysts. Scheme 3 shows the formation mecha-
nism of nitrosobenzene. Taking one carbonyl group of 9,10-an-
thraquinone as an example, the oxygen that has unpaired elec-
trons interacts with the hydrogen atom of a hydrazine mole-
cule. As a result, a hydrogen bond is formed, and the NÀH
bond is weakened. As nitrobenzene approaches the benzene
ring of the catalyst through p–p interactions, the two oxygen
atoms of the nitro group abstract the two activated hydrogen
atoms. Then, the nitro group is reduced to a nitroso group.
After nitrosobenzene is formed, it is converted into aniline di-
rectly through a noncatalytic process. 1,4-Benzoquinone- and
1,4-benzenediol-catalyzed reactions proceed through a similar
Nitrosobenzene was then used as a substrate. After hydra-
zine was added to the nitrosobenzene solution (ethanol was
used as the solvent) at room temperature, the mixture was an-
alyzed immediately by HPLC, and no nitrosobenzene was de-
tected, but the aniline selectivity was 98.6% (Table S2, entry 3).
We further found that nitrosobenzene was also completely
converted and that the same aniline selectivity was observed if
no catalyst was used. In comparison, only 33.7% of hydroxyl-
amine was converted (with an aniline selectivity of 94.3%)
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 4
&
3
&
ÞÞ
These are not the final page numbers!

CHEMCATCHEM
COMMUNICATIONS
www.chemcatchem.org
Science
Foundation
of
China
(21133010,
51221264,
21261160487), the Strategic Priority Research Program of the Chi-
nese Academy of Sciences (Grant No. XDA09030103), and the
Doctoral Starting up Foundation of Liaoning Province, China
(20121068).
Keywords: active sites · carbocatalysis · model catalyst ·
reaction mechanisms · reduction
[1] a) P. Serp, M. Corrias, P. Kalck, Appl. Catal. A 2003, 253, 337–358; b) A.
Schaetz, M. Zeltner, W. J. Stark, ACS Catal. 2012, 2, 1267–1284.
[2] a) D. S. Su, J. Zhang, B. Frank, A. Thomas, X. C. Wang, J. Paraknowitsch,
R. Schlçgl, ChemSusChem 2010, 3, 169–180; b) D. S. Su, S. Perathoner,
G. Centi, Chem. Rev. 2013, 113, 5782–5816.
Scheme 3. Supposed mechanism for the formation of nitrosobenzene cata-
lyzed by 9,10-anthraquinone.
[3] a) J. Zhang, X. Liu, R. Blume, A. H. Zhang, R. Schlçgl, D. S. Su, Science
2008, 322, 73–77; b) B. Frank, J. Zhang, R. Blume, R. Schlçgl, D. S. Su,
Angew. Chem. Int. Ed. 2009, 48, 6913–6917; Angew. Chem. 2009, 121,
7046–7051; c) X. Liu, B. Frank, W. Zhang, T. P. Cotter, R. Schlçgl, D. S. Su,
Angew. Chem. Int. Ed. 2011, 50, 3318–3322; Angew. Chem. 2011, 123,
3376–3380; d) J. Zhang, D. S. Su, R. Blume, R. Schlçgl, R. Wang, X. G.
Yang, A. Gajovic, Angew. Chem. Int. Ed. 2010, 49, 8640–8644; Angew.
Chem. 2010, 122, 8822–8826.
mechanism (Figure S7). Upon using phenanthraquinone as the
catalyst, because the two carbonyl groups are located on the
same side, the nitro group may abstract four hydrogen atoms
at one time and be directly converted into hydroxylamine, but
hydroxylamine is converted very slowly; as a result, a large
amount of hydroxylamine accumulates (Figure S8).
[4] a) D. R. Dreyer, H. P. Jia, C. W. Bielawski, Angew. Chem. Int. Ed. 2010, 49,
6813–6816; Angew. Chem. 2010, 122, 6965–6968; b) D. R. Dreyer, H. P.
Jia, A. D. Todd, J. X. Geng, C. W. Bielawski, Org. Biomol. Chem. 2011, 9,
7292–7295; c) H. P. Jia, D. R. Dreyer, C. W. Bielawski, Adv. Synth. Catal.
2011, 353, 528–532; d) D. R. Dreyer, C. W. Bielawski, Chem. Sci. 2011, 2,
1233–1240; e) H. P. Jia, D. R. Dreyer, C. W. Bielawski, Tetrahedron 2011,
67, 4431–4434; f) D. W. Boukhvalov, D. R. Dreyer, C. W. Bielawski, Y. W.
Son, ChemCatChem 2012, 4, 1844–1849.
[5] Y. Kuang, N. M. Islam, Y. Nabae, T. Hayakawa, M. Kakimoto, Angew.
Chem. Int. Ed. 2010, 49, 436–440; Angew. Chem. 2010, 122, 446–450.
[6] A. Dhakshinamoorthy, M. Alvaro, P. Concepciꢁn, V. Fornꢂs, H. Garcia,
Chem. Commun. 2012, 48, 5443–5445.
[7] a) H. Huang, J. Huang, Y. M. Liu, H. Y. He, Y. Cao, K. N. Fan, Green Chem.
2012, 14, 930–934; b) C. L. Su, M. Acik, K. Takai, J. Lu, S. J. Hao, Y.
Zheng, P. P. Wu, Q. L. Bao, T. Enoki, Y. J. Chabal, K. P. Loh, Nat. Commun.
2012, 3, 1298; c) X. H. Li, M. Antonietti, Angew. Chem. Int. Ed. 2013, 52,
4572–4576; Angew. Chem. 2013, 125, 4670–4674.
In summary, we studied the role of different oxygen func-
tional groups by using a series of model molecules, and the re-
duction of nitrobenzene was selected as a probe reaction. The
carbonyl and hydroxyl groups played important roles, whereas
the carboxylic group had a negative effect. Other oxygenated
groups seemed to be inactive. The reaction occurred most
likely through a direct route, during which nitrosobenzene
could be converted directly into aniline. Considering that the
carbon surface as well as its working mechanism could be
rather complicated, these findings are of some significance for
further investigations of carbon-catalyzed reactions and will
direct the design and synthesis of carbon catalyst for organic
and polymer chemistry.
[8] H. Yu, F. Peng, J. Tan, X. W. Hu, H. J. Wang, J. Yang, W. X. Zheng, Angew.
Chem. Int. Ed. 2011, 50, 3978–3982; Angew. Chem. 2011, 123, 4064–
4068.
Experimental Section
[9] B. H. Han, D. H. Shin, S. Y. Cho, Tetrahedron Lett. 1985, 26, 6233–6234.
[10] B. J. Li, Z. Xu, J. Am. Chem. Soc. 2009, 131, 16380–16382.
[11] Y. J. Gao, D. Ma, C. L. Wang, J. Guan, X. H. Bao, Chem. Commun. 2011,
47, 2432–2434.
[12] C. D. Liang, H. Xie, V. Schwartz, J. Howe, S. Dai, S. H. Overbury, J. Am.
Chem. Soc. 2009, 131, 7735–7741.
[13] J. Zhang, X. Wang, Q. Su, L. J. Zhi, A. Thomas, X. L. Feng, D. S. Su, R.
Schlçgl, K. Mullen, J. Am. Chem. Soc. 2009, 131, 11296–11297.
[14] R. P. Swatloski, S. K. Spear, J. D. Holbrey, R. D. Rogers, J. Am. Chem. Soc.
2002, 124, 4974–4975.
Nitrobenzene reduction
A flask was charged with the catalyst, nitrobenzene (1.2 g), hydra-
zine monohydrate (6.0 equiv.), and ethanol (2 mL). The mixture was
immersed in an oil bath, which was set at 1008C, and the product
was quantified by HPLC.
Characterization
[15] a) Y. Imada, H. Iida, T. Kitagawa, T. Naota, Chem. Eur. J. 2011, 17, 5908–
5920; b) A. Dhakshinamoorthy, S. Navalon, D. Sempere, M. Alvaro, H.
Garcia, Chem. Commun. 2013, 49, 2359–2361; c) W. M. N. Ratnayake,
J. S. Grossert, R. G. Ackman, J. Am. Oil Chem. Soc. 1990, 67, 940–946.
[16] A. Corma, P. Concepcion, P. Serna, Angew. Chem. Int. Ed. 2007, 46,
7266–7269; Angew. Chem. 2007, 119, 7404–7407.
Characterization by NMR spectroscopy was performed with
a Bruker 400 MHz NMR spectrometer with tetramethylsilane as an
internal reference. UV detection was performed with a Cary 5000
UV/Vis/NIR spectrophotometer.
[17] F. Visentin, G. Puxty, O. M. Kut, K. Hungerbuehler, Ind.Eng. Chem. Res.
2006, 45, 4544–4553.
Acknowledgements
The authors acknowledge the financial support from the Ministry
of Science and Technology (2011CBA00504), the National Natural
Received: February 23, 2014
Published online on && &&, 0000
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 4
&
4
&
ÞÞ
These are not the final page numbers!

COMMUNICATIONS
Modeling carbon: Thanks to model
molecules, the carbon-catalyzed reduc-
tion of nitrobenzene is mimicked. The
role of different oxygen functional
groups on a carbon catalyst is studied,
and the carbonyl and hydroxyl groups
seem to be the most important moiet-
ies, which may be ascribed to their abili-
ty to activate hydrazine. The reaction
occurs more likely through a direct
route, during which nitrosobenzene
may also be converted directly into
aniline.
S. Wu, G. Wen, X. Liu, B. Zhong, D. S. Su*
&& – &&
Model Molecules with Oxygenated
Groups Catalyze the Reduction of
Nitrobenzene: Insight into
Carbocatalysis
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 0000, 00, 1 – 4
&
5
&
ÞÞ
These are not the final page numbers!