Preparation and catalytic performance of carbon nanotube supported palladium catalyst

Article abstract of DOI:10.1002/cjoc.201090165

Two carbon nanotube supported palladium catalysts were prepared using a chemical reduction technique (Pd/CR-CNT) and a conventional impregnation method (Pd/CNT) respectively, and their catalytic performances for Heck reaction were investigated. The catalysts were characterized by TEM and XPS techniques and the products were characterized by ¹H NMR. Research results showed that the (Pd/CR-CNT) catalyst showed a better catalytic activity than the (Pd/CNT) catalyst, owing to better dispersion of palladium nanoparticles and stronger interaction between the active palladium species and carbon nanotube. Meanwhile, the product yield maintained 99.93% of its initial value at five-times re-use, compared with that at the first time use. The catalyst prepared with the chemical reduction method represented a better reusing performance.

Full text of DOI:10.1002/cjoc.201090165

FULL PAPER
Preparation and Catalytic Performance of Carbon Nanotube
Supported Palladium Catalyst
Zhang, Yan(张艳)
Chu, Wei*(储伟)
Xie, Lijuan(谢丽娟)
Sun, Wenjing(孙文晶)
College of Chemical Engineering, Sichuan University, Chengdu, Sichuan 610065, China
Two carbon nanotube supported palladium catalysts were prepared using a chemical reduction technique
(Pd/CR-CNT) and a conventional impregnation method (Pd/CNT) respectively, and their catalytic performances for
Heck reaction were investigated. The catalysts were characterized by TEM and XPS techniques and the products
1
were characterized by H NMR. Research results showed that the (Pd/CR-CNT) catalyst showed a better catalytic
activity than the (Pd/CNT) catalyst, owing to better dispersion of palladium nanoparticles and stronger interaction
between the active palladium species and carbon nanotube. Meanwhile, the product yield maintained 99.93% of its
initial value at five-times re-use, compared with that at the first time use. The catalyst prepared with the chemical
reduction method represented a better reusing performance.
Keywords carbon nanotube (CNT), supported catalyst, heck reaction, chemical reduction
Introduction
composed (100%) at 723 K, and the catalytic activity of
Rh/MWNT (Multi-walled carbon Nanotube) was higher
than that of Rh/γ-Al2O3. XPS results showed that carbon
nanotube could stabilize the catalyst active sites (the
zero-valent Rh) compared with the γ-Al2O3 support,
therefore, Rh/MWNT displayed a higher catalytic per-
formance in NO decomposition reaction. It was gener-
ally believed that the smaller the metal particle size, the
more susceptibly the substrate will be oxidized. But the
Rh particle size was 10.7 nm in 1 wt% Rh/MWNT
catalyst, much smaller than 25.7 nm in 1 wt%
Rh/γ-Al2O3. This attributed to the strong interaction
between metal clusters and the MWNT, which made
zero-valent Rh with smaller particle size and more sta-
ble on the surface of MWNT. The metal-supporter
strong interaction (SMSI) effect was more significant in
carbon nanotube supported metal catalysts than that in
other carbon materials.22, 23
However, in the preparation of catalysts, the palla-
dium nanoparticles were easy to agglomerate. The addi-
tion of chemical reductant could promote the dispersion
of palladium nanoparticles and strengthen the interac-
tion of active species with support. In this paper, CNT
was used as the carrier materials and two carbon nano-
tube supported palladium catalysts were prepared using
chemical reduction and conventional impregnation
methods respectively. By investigating their catalytic
performances in Heck reaction with the assistance of
XPS and TEM characterization, the reason that chemical
reduction can improve catalytic activity and stability in
the Heck reaction was investigated.
Heck reaction was a general term for the synthesis re-
actions of styrene-based compounds, through catalyzing
aryl halide with carbon-carbon double bond compounds,
in the presence of transition metal catalysts.1,2 It was a
useful method for forming C—C bond on the double
bond carbon atoms, and it has found with the widely
application to organic synthesis.3-7 In the catalysts of
Heck reaction, precious metal palladium was the most
used active component. Compared with the traditional
homogeneous Pd(OAc)2, PdCl2 catalysts in Heck reac-
tion, supported palladium catalysts possessed many ad-
vantages, such as high catalytic activity, good stability,
easy separation and satisfactory reusability, which have
been favored by researchers at home and abroad.
Though attempts have been made to achieve higher
performances by effectively dispersing Pd precursors
and Pd-based nanoparticles onto the support materials,
more effort is still necessary.8-12 Among the possible
supports, carbon black has been widely used to disperse
Pd nanoparticles as traditional carbon support materi-
als.13 Carbon nanotubes (CNT) were considered to be a
more attractive candidate, owing to their outstanding
mechanical characteristics, good electronic conductivity,
nanometer size and high accessible surface area, which
were also of great interest for many applications to, for
example, batteries, flat panel displays, chemical sensors
and fuel cells.14-20
The application of carbon nanotube catalysis was a
hot topic in recent years. Luo et al.21 loaded 1 wt% Rh
on carbon nanotube and found that NO was fully de-
*
E-mail: chuwei65@yahoo.com.cn; Tel.: 0086-028-85403836; Fax: 0086-028-85461108
Received May 10, 2009; revised October 29, 2009; accepted January 14, 2010.
Project supported by Program for New Century Excellent Talents (No. NCET-05-0783) and the Natioinal Natural Science Foundation of China (No.
20776089).
Chin. J. Chem. 2010, 28, 879— 883
© 2010 SIOC, CAS, Shanghai, & WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
879

Zhang et al.
FULL PAPER
Experimental
Materials
the reaction was completed, added 30 mL of deionized
water and 1.0 g of Na2CO3, stirred for 20 min, then the
catalyst was separated by filtration, and reused after
washing and drying. The filtrate separated from the re-
action mixture was treated directly with 2% HCl solu-
tion (40 mL) to precipitate crude products, which were
purified by recrystallization to give isolated yield.
Carbon nanotubes were produced via a chemical va-
por deposition (CVD) method, and the purity was about
95%, obtained from the Organic Chemical Limited
Company, Chengdu, China. The raw material was func-
tionalized by refluxing in concentrated HNO3 (65 wt%)
for 3 h at 140 ℃ in an oil bath, as previously re-
ported.24 Iodobenzene (98% from Aldrich), palladium
chloride ( ＞ 99.9%, A.R.), N,N-dimethylformamide
(DMF, 99.5%, A.R.) and triethylamine (98%, A.R.)
were used without further purification.
Recycling test
After the reaction of iodobenzene and acrylic acid
(DMF as solvent, triethylamine as base, at 100 ℃,
100% conversion of iodobenzene for each cycle), the
catalyst was separated by filtration, washed with ethanol
(1 mL×3) and deionized water (2 mL×3) sequentially,
and dried at 80 ℃ under vacuum for 8 h, then reused in
Heck coupling.
Instruments
1H NMR (Bruker Avance II-400 MHz), XPS (V4105,
C1s＝284.6 eV) and TEM (JEOL-JEM-200CX), were
employed to characterize the catalytic materials.
Results and discussion
Preparation of Pd/CNT and Pd/CR-CNT catalysts
Pd/CNT catalyzed Heck reaction of different sub-
strates
The reaction sequence for the dispersion of Pd
nanoparticles on the CNT surface is depicted as follows:
(a) Catalyst prepared by a conventional impregnation
method. 1.0 g of purified carbon nanotube was added
into a certain amount of ethanol solution of palladium
chloride (800 mg/L), and the mixture was magnetically
stirred for 10 h at 60 ℃. After cooling down, the com-
pound was filtered and washed with deionized water,
and dried in vacuum for 12 h at 80 ℃ to give the cata-
lyst (recorded as Pd/CNT).
Carbon nanotube supported palladium catalyst could
be a good catalyst for the reactions of iodobenzene with
vinyl compounds. In this paper, the acrylic acid, styrene,
methyl acrylate and ethyl acrylate were selected to react
with iodobenzene (Scheme 1). When x(Pd) was equal to
0.026%, under ambient atmosphere, high yields of
trans-cinnamic acid, stilbene, methyl cinnamate, and
1
ethyl cinnamate were achieved (Table 1). The H NMR
spectra data confirmed the product structures by com-
parison with standard structures (Table 2).
(b) Catalyst prepared by a chemical reduction
method. Sodium thiosulfate as a chemical reductant was
configured to solution concentration of 5 mol/L. 1.0 g of
purified carbon nanotube was taken into a specified
volume of ethanol solution of palladium chloride (800
mg/L) and the corresponding amount of 5 mol/L sodium
thiosulfate solution was added in the continuously stir-
ring condition. The mixture was reacted for about 10 h
at 60 ℃, then filtered and washed with deionized water
and dried in vacuum for 12 h at 80 ℃ to afford the
catalyst (recorded as Pd/CR-CNT).
Catalytic performance of Pd/CNT and Pd/CR-CNT
catalysts in the Heck reaction
The carbon nanotube supported Pd/CNT, Pd/CR-
CNT catalysts were applied to the reactions of iodoben-
zene, bromobenzene with acrylic acid under ambient
atmosphere, the reaction of iodobenzene with acrylic
acid should proceed at 100 ℃ and the reaction of bro-
mobenzene with acrylic acid should proceed at 120 ℃.
Table 3 indicated that Pd/CR-CNT displayed a better
catalytic activity, which was due to the addition of
chemical reductant. The addition of chemical reductant
General procedure for Heck reaction
In a 100 mL round-bottomed flask, the mixture of
aryl iodide (6.0 mmol), vinyl compounds (6.6 mmol),
triethylamine (6.6 mmol, the substrate of acrylic acid
needed triethylamine 13.2 mmol), and a certain amount
of carbon nanotube supported palladium catalyst, and
DMF (5 mL) was stirred under ambient atmosphere at
100 ℃. The reaction was monitored by TLC.25,26 When
Scheme 1
Table 1 Heck arylation catalyzed by the CNT supported palladium catalyst
Reactant
Reaction time/h
Product
Yield/%
70.3
Iodobenzene＋acrylic acid
Iodobenzene＋styrene
Iodobenzene＋methyl acrylate
Iodobenzene＋ethyl acrylate
5
5
8
8
Cinnamic acid
Stilbene
53.6
Methyl cinnamate
Ethyl cinnamate
57.6
58.2
880
www.cjc.wiley-vch.de
© 2010 SIOC, CAS, Shanghai, & WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Chin. J. Chem. 2010, 28, 879— 883

Preparation and Catalytic Performance of Carbon Nanotube Supported Palladium Catalyst
Table 2 1H NMR data of products 2a—2d
No.
2a
1H NMR, δ (CDCl3, 400 MHz)
6.49 (d, J＝16.0 Hz, 1H), 7.21—7.58 (m, 5H), 7.82 (d, J＝16.0, 1H)
2b
2c
7.14 (s, 2H), 7.29—7.56 (m, 10H)
3.81 (d, J＝7.1 Hz, 3H), 6.44 (d, J＝16.0 Hz, 1H), 7.21—7.58 (m, 5H), 7.69 (d, J＝16.0 Hz, 1H)
1.34 (t, J＝7.1 Hz, 3H), 4.26 (q, J＝7.1 Hz, 2H), 6.44 (d, J＝16.0 Hz, 1H), 7.21—7.58 (m, 5H), 7.69 (d, J＝16.0 Hz, 1H)
2d
Table 3 Catalytic performance of the CNT supported palladium catalysts in Heck reaction
Reactant
Reaction time/h
3
Product
Catalyst
Pd/CNT
Yield/%
70.3
Iodobenzene＋acrylic acid
Cinnamic acid
Pd/CR-CNT
Pd/CNT
70.7
0.0
Bromobenzene＋acrylic acid
8
Cinnamic acid
Pd/CR-CNT
20.1
Morphology analysis of Pd/CNT and Pd/CR-CNT
catalysts
could promote the dispersion of palladium nanoparticles
for the deep reduction of palladium species.
The effect of catalyst amount on the catalytic per-
formance was gathered in Table 4, where x(Pd) repre-
sents the mole ratio of Pd to iodobenzene. Iodobenzene
and acrylic acid were reacted under ambient atmosphere
at 100 ℃. When x(Pd) was equal to 0.004%, the reac-
tion could proceed smoothly and the yield of cinnamic
acid was 62.5%. With the increase in x(Pd), reaction
time was shortened and the yield was promoted. How-
ever, when x(Pd) increased to 0.026%, the yield and
reaction time were no longer changed significantly fol-
lowing the increase of x(Pd).
The transmission electron microscopy (TEM) images
for the two carbon nanotube supported palladium cata-
lysts were showed in Figure 1. A conspicuous agglom-
eration can be seen in the Pd/CNT (a) as a result of
fewer surface defects on the carbon nanotube. To re-
solve the problem, the chemical reduction method was
adopted. The improved results can be drawn from the
picture (b): the palladium nanoparticles were evenly
dispersed on the surface of carbon nanotube, with the
partical size about 10—20 nm in Pd/CNT, which was
mainly attributed to the function on the port and the lat-
eral wall of carbon nanotube, when the carbon nanotube
were treated with concentrated nitric acid. Moreover, the
addition of chemical reductant could promote the dis-
persion of palladium nanoparticles and strengthen the
interaction of active species with support, leading to the
Table 4 Effect of x(Pd) on the catalytic performance
x(Pd)/%
0.125
0.052
0.026
0.015
0.004
Reaction time/h
Yield/%
70.8
5
5
70.6
5
70.3
8
66.7
10
62.5
Performance of recycling use
The catalysts could be reused by simple filtration,
followed by washing with ethanol and deionized water.
The Pd/CNT and Pd/CR-CNT catalysts were able to be
reused 5 times in DMF and did not break down mechani-
cally and/or chemically in recycling. The Pd/CR-CNT
showed higher stability than Pd/CNT. In the recycling use
of Pd/CR-CNT, the product yield maintained 99.93% of
that in the first time (70.7%). This also showed that the
addition of chemical reductant was beneficial to strength-
ening the interaction of active species with support.
Table 5 Relationship between the recycling times of catalyst
and product yield
Recycling time
Catalyst
1
2
3
4
5
Figure 1 TEM images of (a) Pd/CNT and (b) Pd/CR-CNT
catalysts.
Pd/CNT
70.3
70.7
65.9
69.9
64.2
68.4
63.9
67.7
63.1
65.4
Pd/CR-CNT
Chin. J. Chem. 2010, 28, 879— 883
© 2010 SIOC, CAS, Shanghai, & WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.cjc.wiley-vch.de
881

Zhang et al.
FULL PAPER
active metal components more easily uniformly dis-
persed on the surface of to be carbon nanotube.24,27 So,
the carbon nanotube supported palladium catalyst pre-
pared by chemical reduction possessed better dispersion
and smaller particle size.
showed a better reusing performance, the product yield
maintained 99.93% at five-times reusing, compared with
that at the first time using. At the same time, the
Pd/CR-CNT catalyst possessed a good catalytic activity
for the reactions of iodobenzene with acrylic acid, sty-
rene, methyl acrylate and ethyl acrylate. All the im-
proved performances are determined by the addition of
reductant in the preparing process of catalyst, because
adding reductant promotes the dispersion of palladium
nanoparticles and part palladium species transformed
into zero-valent palladium atoms, which strengthen the
interaction of active species with carbon nanotube.
Surface analysis of Pd/CNT and Pd/CR-CNT cata-
lysts
The X-ray photoelectron spectroscopy (XPS) char-
acterization data of Pd, PdCl2, CNT and the two carbon
nanotube supported palladium catalysts with different
preparing methods were gathered in Table 6. In the two
catalysts, the binding energy for Pd3d5/2 was equal to
335.8 and 335.5 eV, responding to zero-valent palla-
dium atoms, and the other binding energy for Pd3d5/2 was
equal to 337.3 and 337.1 eV, responding to divalent
palladium cations.28 The two binding energies for Pd3d5/2
in the catalysts were also observed, namely the palla-
dium species possessed two types of valent states which
signified that part of the divalent palladium cations had
been transformed to the zero-valent palladium atoms.
335.8 eV (335.5 eV) was very close to the zero-valent
palladium atoms (335.3 eV) and 337.3 eV (337.1eV)
approached to divalent palladium cations (338.3 eV). At
the same time, it was observed that the binding energies
of Pd3d5/2 in two catalysts had significant shift compared
with zero-valent palladium atoms and divalent palla-
dium cations, which was attributed to the interaction of
palladium species with the carbon nanotube.29 By the
way, the offset amount of binding energy in the
Pd/CR-CNT catalyst was larger than that in the Pd/CNT
catalyst, attributed to the stronger interaction between
active species and carbon nanotube. The Pd/CR-CNT
catalyst had higher relative content of zero-valent palla-
dium on the catalyst surface, which also illustrated that
the addition of chemical reductant was beneficial to
strengthening the interaction of active species with sup-
port, and the carbon nanotube could stabilize the
zero-valent palladium atoms.21
References
1
Littke, A. F.; Fu, G. C. Angew Chem., Int. Ed. 2002, 41,
4176.
2
3
4
Heck, R. F. Org. Reac. 1982, 27, 345.
Bedford, R. B. Chem. Commun. 2003, 1787.
Amengual, R.; Genin, E.; Michelet, V. Adv. Synth. Catal.
2002, 344, 393.
5
6
Yang, N. C.; Such, D. H. Polym. Bull. 2001, 46, 29.
Wang, Z. L.; Wang, L.; Yan, J.-C. Chin. J. Chem. 2008, 26,
1721.
7
8
9
Wang, J.; Xu, S.; Tao, X.-C. Chin. J. Org. Chem. 2008, 28,
52 (in Chinese).
Wang, Z.; Zhu, Z.-Z.; Li, Y.-X.; Li, H.-L. Chin. J. Chem.
2008, 26, 666.
Avelino, C.; Hermenegildo, G.; Antonio, L. J. Mol. Catal. A:
Chem. 2005, 230, 97.
10 Chen, X.-C.; Hou, Y.-Q.; Wang, H. J. Phys. Chem. C 2008,
112, 8172.
11 Wang, S.-J.; Yin, S.-F.; Li, L. Appl. Catal. B 2004, 52, 287.
12 Wang, W. D.; Serp, P.; Kalck, P. J. Mol. Catal. A, Chem.
2005, 235, 194.
13 Ha, S.; Larsen, R.; Masel, R. I. J. Power Sources 2005, 144,
28.
14 Iijima, S.; Ichihashi, T. Nature 1993, 363, 603.
15 Wong, E. W.; Sheehan, P. E.; Lieber, C. M. Science 1997,
277, 1971.
16 Che, G.; Lakshmi, B. B.; Martin, C. R.; Fisher, E. R. Lang-
muir 1999, 15, 750.
17 Wilder, J. W. G.; Venema, L. C.; Rinzler, A. G.; Smalley, R.
E.; Dekker, C. Nature 1998, 391, 59.
18 Kong, J.; Franklin, N. R.; Zhou, C.; Chapline, M. G.; Peng,
S.; Cho, K.; Dai, H. Science 2000, 287, 622.
19 Chen, J.; Wang, M.; Liu, B.; Fan, Z.; Cui, K.; Kuang, Y. J.
Phys. Chem. B 2006, 110, 11775.
20 Huang, L.-H.; Chu, W.; Hong, J.-P.; Luo, S.-Z. Chin. J.
Catal. 2006, 27, 596 (in Chinese).
Table 6 XPS data of Pd/CNT and Pd/CR-CNT (eV)
＋
Sample
Pd3d5/2
Intensity ratio (Pd0/Pd2 )
O1s
Pd/CNT
335.8 (337.3)
0.582
0.672
532.73
532.71
Pd/CR-CNT 335.5 (337.1)
All relative to C1s＝284.6 eV, Pd (Pd3d5/2)＝335.3 eV, PdCl2
(Pd3d5/2)＝338.3 eV, CNT (O1s)＝532.76 eV.
Conclusion
In this work, a chemical reduction method and a tra-
ditional impregnation method were introduced into pre-
paring carbon nanotube supported palladium catalyst.
Research results show that the Pd/CR-CNT catalyst has
a better catalytic activity and higher stability. The
Pd/CR-CNT catalyst could catalyze the Heck reaction of
bromobenzene, and the product yield reached 20.1%,
while the Pd/CNT catalyst could not. Meanwhile, the
catalyst prepared with the chemical reduction method
21 Luo, J. Z.; Gao, L. Z.; Leung, Y. L. Catal. Lett. 2000, 66, 91.
22 Liu, Z. Q.; Ma, J. Acta Chim. Sinica 2007, 65, 2965 (in Chi-
nese).
23 Wang, Z.; Zhu, Z.-Z.; Li, H.-L Acta Chim. Sinica 2007, 65,
1149 (in Chinese).
24 Zhang, X.-W.; Chu, W.; Zhuang, H.-X.; Xu, S.-W. Chem. J.
Chin. Univ. 2005, 26, 493 (in Chinese).
25 Yang, Y.-F.; Zhuang, M.; Zeng, C.-X.; Huang, C.-B.; Luo,
882
www.cjc.wiley-vch.de
© 2010 SIOC, CAS, Shanghai, & WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Chin. J. Chem. 2010, 28, 879— 883

Preparation and Catalytic Performance of Carbon Nanotube Supported Palladium Catalyst
M.-F. Chin. J. Chem. 2006, 24, 1309.
26 Biffis, A.; Zecca, M.; Basato, M. J. Mol. Catal. A: Chem.
2001, 173, 249.
27 Mehnert, C. P.; Weaver, D. W.; Ying, J. Y. J. Am. Chem. Soc.
1998, 120, 12289.
28 Tangbao, H.-J.; Lü, R.-Q.; Xiang, S. H. Acta Petrolei Sinica
(Petroleum Processing Section) 2007, 23, 13 (in Chinese).
29 Khodakov, A. Y.; Chu, W.; Fongarland, P. Chem. Rev. 2007,
107, 1692.
(E0905101 Sun, H.; Zheng, G.)
Chin. J. Chem. 2010, 28, 879— 883
© 2010 SIOC, CAS, Shanghai, & WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.cjc.wiley-vch.de
883