Synthesis of modified carbon nanotube-supported Pd and the catalytic performance for hydrodehalogenation of aryl halides

Article abstract of DOI:10.1139/cjc-2012-0414

The effects of surface functionalization on the deposition of Pd nanoparticles on multi-walled carbon nanotubes (CNTs) have been studied. Two methods were used for the functionalization of CNTs: one employed a 1:1 (v/v) mixture of concentrated H₂SO₄/HNO₃, sonicated at RT; the other employed a 3:1 (v/v) mixture of concentrated H ₂SO₄/HNO₃, sonicated at 60 C. A large number of surface oxide groups were introduced on the surface of functionalized CNTs, especially in strong oxidation power (H₂SO₄/HNO ₃ = 3:1, 60 C). The dispersion of Pd nanoparticles on CNTs was found to depend on the amount of surface oxide groups, with larger amounts of surface oxide groups resulting in higher dispersion of Pd nanoparticles. The effects of CNT surface functionalization on the performance of Pd/CNTs catalysts were studied, and the reaction conditions optimized, using the hydrodebromination of bromobenzene as model reaction. Under optimal conditions, the hydrodehalogenation of various aryl halides were tested over Pd/CNTs catalysts.

Full text of DOI:10.1139/cjc-2012-0414

307
ARTICLE
Synthesis of modified carbon nanotube-supported Pd and the catalytic
performance for hydrodehalogenation of aryl halides
Yuandong Xu, Long Chen, and Xiaolai Wang
Abstract: The effects of surface functionalization on the deposition of Pd nanoparticles on multi-walled carbon nanotubes
(CNTs) have been studied. Two methods were used for the functionalization of CNTs: one employed a 1:1 (v/v) mixture of
concentrated H2SO4/HNO3, sonicated at RT; the other employed a 3:1 (v/v) mixture of concentrated H2SO4/HNO3, sonicated at
60 °C. A large number of surface oxide groups were introduced on the surface of functionalized CNTs, especially in strong
oxidation power (H2SO4/HNO3 = 3:1, 60 °C). The dispersion of Pd nanoparticles on CNTs was found to depend on the amount of
surface oxide groups, with larger amounts of surface oxide groups resulting in higher dispersion of Pd nanoparticles. The effects
of CNT surface functionalization on the performance of Pd/CNTs catalysts were studied, and the reaction conditions optimized,
using the hydrodebromination of bromobenzene as model reaction. Under optimal conditions, the hydrodehalogenation of
various aryl halides were tested over Pd/CNTs catalysts.
Key words: CNTs, functionalization, hydrodehalogenation, aryl halides.
Résumé : On a étudié les effets de la fonctionnalisation de la surface sur le dépôt de nanoparticules des nanotubes de carbone
(NTC) a` multiples parois. Deux méthodes ont été utilisées pour la fonctionnalisation des NTC: la première fait appel a` un mélange
1:1 (v/v) d'acides concentrés H2SO4/HNO3 et une sonification a` la température ambiante alors que l'autre utilise un mélange 3:1 (v/v)
d'acides concentrés H2SO4/HNO3 et une sonification a` 60 °C. On a pu ainsi introduire un grand nombre de groupes oxydes de
surface sur la surface de NTC fonctionnalisées, particulièrement avec le grand pouvoir d'oxydation du mélange 3:1 H2SO4/HNO3,
a` 60 °C. On a trouvé que la dispersion des nanoparticules de Pd sur les NTC dépend de la quantité de groupes oxydes de surface;
la présence de plus grandes quantités de groupes oxydes de surface conduisant a` une plus grande dispersion des nanoparticules
de Pd. On a aussi étudié les effets de la fonctionnalisation de la surface des NTC sur la performance des catalyseurs Pd/NTC et on
a optimisé les conditions réactionnelles en utilisant la réaction d'élimination de HBr du bromobenzène comme réaction modèle.
Les conditions optimales ainsi trouvées, ont été évaluées pour l'élimination de HX (déshydrohalogénation) a` partir de divers
halogénures d'aryles, sur des catalyseurs Pd/NTC. [Traduit par la Rédaction]
Mots-clés : nanoparticules des nanotubes de carbone (NTC) a` multiples parois, élimination de HX (déshydrohalogénation),
halogénures d'aryles.
also an important method for chemical conversion in organic
Introduction
synthesis.4
Aryl halides are widely used as solvents, perfume, insect repel-
So far, a variety of catalysts for the dehalogenation reactions
lent, fungicidal agents, and organic intermediates of many syn-
with CTH method have been developed, including the homoge-
thesis processes. The activity of the C–X (X = F, Cl, Br, I, etc.) bond
neous Pd complexes,5 Pd/C,6 Pd/ZrO2,7 Pd/Al2O3,8 and other metal
in aryl halides is usually lower, even under a strong reaction con-
oxides and amphoteric polystyrene-poly (ethylene glycol) resin-
dition, than in halogenated alkanes. As a result, it is rather diffi-
supported Pd catalyst.9 While these catalysts exhibit good perfor-
cult for aryl halides to be degraded once they are released into the
mance in the dehalogenation reactions, some problems remain.
environment. Actually, aryl halides have been recognized as one
Generally, these problems include: (i) the high activity of catalysts
usually relies on the high Pd loadings (5ϳ10 wt%), so the actual
cost of catalysts is very high; (ii) the H–X acids generated during
the reaction poison the active sites of the catalysts, resulting in
reduced stability and lifetime of the catalysts; (iii) the loss of Pd in
liquid phase reactions is serious, and the separation and recycling
of catalysts are difficult to operate. Since these problems are gen-
erally associated with catalysts, many researchers believe that a
solution to these problems is to develop some novel catalysts.10,11
In recent years, the application of carbon nanotubes (CNTs) in
the catalysis field has attracted considerable attention.12 Besides
the advantages of traditional carbon materials, (e.g., resistance to
acid and basic medium, and the adjustment of pore structure and
of the most widespread and persistent toxic pollutants, which
directly threaten human survival and development.1 The disposal
of these organic wastes is a vital environmental issue. Several
methods, such as physical, biological, and chemical oxidation,
have been developed to reduce the toxicity of aryl halides. How-
ever, each method has its own drawbacks, i.e., high cost, high
energy consumption, and incomplete pollutant elimination,
which largely limit their application and development.2 Recently,
a catalytic transfer hydrogenation (CTH) method for the degrada-
tion of aryl halides has received much attention. This method not
only has low reaction temperature and energy consumption but
effectively prohibits the formation of more toxic compounds such
as dioxins.3 In addition to environmental remediation, CTH is
Received 25 October 2012. Accepted 18 December 2012.
Y. Xu. School of Chemistry and Chemical Engineering, Henan University of Technology, Zhengzhou 450001, P.R. China; State Key Laboratory for Oxo Synthesis and
Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, P.R. China.
L. Chen and X. Wang. State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou
730000, P.R. China.
Corresponding authors: Long Chen (e-mail: chenlong2000_1984@126.com) and Xiaolai Wang (e-mail: xlwang@licp.cas.cn).
Can. J. Chem. 91: 307–314 (2013) dx.doi.org/10.1139/cjc-2012-0414
Published at www.nrcresearchpress.com/cjc on 20 December 2012.

308
Can. J. Chem. Vol. 91, 2013
surface properties), CNTs also possess a unique hollow structure,
unique electronic, mechanical, and adsorption properties, as well
as excellent thermal stability. Therefore, CNTs are very promising
candidates for catalyst supports. Because of their small sizes, CNT-
supported catalysts can be uniformly dispersed into the solvents
under stirring.13 Moreover, the catalysts can be recovered via gravity
settling or filtration.13,14 Most importantly, for liquid phase reactions
the high external surface area and the lack of microporous structure
of CNTs could remarkably reduce the mass transfer phenomena dur-
ing the reactions, which greatly improve catalytic efficacy. The use of
CNTs as catalyst supports in many reactions has been reported, and
higher activity and (or) selectivity were observed on CNT-supported
catalysts compared with catalysts supported on conventional
carriers.13–15 However, the application of CNTs in catalytic
hydrodehalogenation of aryl halides is less well-known.
In a previous publication, we reported for the first time that com-
pared with conventional supports, CNT-supported Pd catalysts offer
not only better performance in the hydrodehalogenation of
bromobenzene, but also a substantial reduction in the amount of
Pd required.16 Inspired by this study, Kim et al. reported the syn-
thesis of Pd/CNT nanocomposites and investigated their catalytic
behavior in the hydrodehalogenation of aryl halides.17 In this
work, we first studied the effects of surface functionalization of
CNTs and reaction conditions on the performance of Pd/CNT cat-
alysts in the liquid phase hydrodehalogenation of bromobenzene.
Subsequently, we extended the substrate to various aryl halides
over Pd/CNT catalysts under optimal reaction conditions.
Cyclic voltammetry (CV) measurements were employed to quali-
tatively determine the number of oxygen-containing functional
groups on the surface of the CNTs, using a CHI 660 C electrochem-
ical workstation (CHI Instrument Corp., Shanghai). The glassy car-
bon electrode where 5 ␮L of the suspension (1 g/L) of CNTs in DMF
was dropped and subsequently dried by IR lamp was used as the
working electrode; platinum wire and saturated calomel elec-
trodes were used as counter and reference electrodes, respec-
tively. The electrolyte was 1 mol/L H2SO4 solution.
Catalytic performance test
Hydrodehalogenation of the aryl halides was carried out in a
glass flask equipped with a reflux condenser under atmospheric
pressure. Typically, at a certain temperature, 50 mg of catalyst,
hydrazine hydrate (85%), solvent, and reaction substrate were
mixed and stirred for different times. The products were quanti-
tatively analyzed by gas chromatography (GC7890II, Shanghai
Tian Mei Scientific Instrument Co., Ltd.) and toluene was used as
internal standard. The products were further confirmed by
GC–MS (Agilent 6890N/5973N).
Results and Discussions
Modification of CNTs
Figure 1 shows the C1s and O1s core level spectra of r-CNTs,
f-CNTs-1, and f-CNTs-2. Negligible difference in C1s spectra is de-
tected between crude and acid-treated CNTs (Fig. 1a), indicating
that the structure of CNTs was not seriously destroyed during acid
treatment. In contrast, the O1s spectra show a remarkable differ-
ence (Fig. 1b). The 531.6 eV and 533.0 eV peaks in the spectra can be
assigned to the O element in –C=O and –C–O functional groups,
respectively19,20 The 534.6 eV peak can be attributed to the O
element in –COO– functional groups.19 It is noted that the surface-
adsorbed H2O in the samples may also contribute to this peak,
since the binding energies for H2O and –COO– are found to over-
lap.20 The number of the surface oxygen-containing functional
groups on CNTs can be qualitatively determined by the relative
intensity of the O1s spectra. The relatively weak peaks for r-CNTs
demonstrate that there were a small amount of oxygen-containing
functional groups on their surfaces, which are likely to be produced
during the chemical vapor deposition process for CNTs prepara-
tion.21 In contrast, the relatively stronger peaks for f-CNTs-1 and
f-CNTs-2 suggest that a large number of –C=O, –OH, and –COO– func-
tional groups were introduced on their surfaces during the acid treat-
ment. Moreover, the number of oxygen-containing functional
groups depends on the acid treatment conditions, with a harsher
condition resulting in a larger number of surface functional groups.
Table 1 lists the C and O atomic percentages determined by XPS for
various CNT samples. Consistent with Fig. 1b, the O content in the
samples follows the order f-CNTs-2 > f-CNTs-1 > r-CNTs.
To further confirm the surface functional groups and their den-
sities on the CNT supports, we carried out electrochemical char-
acterizations. Figure 2 shows the CV curves of the r-CNTs, f-CNTs-1,
and f-CNTs-2 modified electrodes in 1 mol/L H2SO4. For the modi-
fied CNTs (f-CNTs-1 and f-CNTs-2), the ϳ0.55 V anodic peak and the
ϳ0.49 V cathodic peak in CV curves are related to the redox phe-
nomena of oxygen-containing functional groups. Almost no redox
peaks were observed for r-CNTs, demonstrating that there were
few functional groups on their surfaces. The intensity of the redox
peaks of f-CNTs-2 was greater than that of f-CNTs-1, suggesting
that more oxygen-containing functional groups were introduced
on the former than on the latter. The electrochemical characteriza-
tion results are in good agreement with the XPS measurements.
The properties and content of the surface functional groups on
CNTs were further characterized by FTIR. Figure 3 shows the FTIR
spectra of r-CNTs, f-CNTs-1, and f-CNTs-2. The peak at 1562 cm−1 is
the characteristic band of aromatic ring.22 Since CNTs is com-
posed of rings of 5 carbon atoms and 6 carbon atoms, similar to
Experimental
Modification of CNTs supports
High quality multi-walled CNTs were synthesized according to a
method described previously.18 Raw CNTs (r-CNTs) were treated by
a mixture of concentrated H2SO4 and HNO3 in ultrasonication
conditions. Briefly, 1.0 g of r-CNTs was added to a 100 mL round-
bottomed flask containing 1:1 (v/v) of the concentrated H2SO4 and
HNO3 mixture. The flask was placed in an ultrasonic bath for ϳ4 h
at room temperature (RT). The modified CNTs were then filtered,
washed with water to neutral, dried at 100 °C overnight, and de-
noted as f-CNTs–1. The r-CNTs were also treated in a harsher con-
dition (3:1 v/v) of the concentrated H2SO4 and HNO3 mixture, and
at an ultrasonic bath temperature of 60 °C). The products obtained
were designated as f-CNTs-2.
Catalyst preparation
Pd/CNTs catalysts were prepared via impregnation with a nom-
inal Pd loading of 2 wt%. Specifically, the CNTs (r-CNTs, f-CNTs-1, or
f-CNTs-2) were first impregnated into a PdCl2 aqueous solution for
30 min. The mixture was then treated in ultrasonic bath at RT
for ϳ4 h, dried at 100 °C for 12 h, and calcined at 300 °C for 3 h.
Finally, the samples were reduced at 250 °C under hydrogen at-
mosphere for 2 h and cooled to RT under the protection of Ar flow.
Catalyst characterization
The surface composition and properties of samples were ana-
lyzed by X-ray photoelectron spectroscopy (XPS, VG ESCALAB
210 type). The base pressure of the main vacuum chamber was
4.5 × 10−11 mbar (1 bar = 100 kPa) and the X-ray source was Mg
K␣ (h␷ = 1253.6 eV). The charge correction was based on C1s bind-
ing energy (BE, 284.4 eV). The X-ray diffraction (XRD) analysis of
the catalysts were performed using a PANalytical X'Pert PRO X,
with K␣ radiation (␭ = 1.542 Å), generated at 50 kV, 60 mA. A
Thermo Nicolet 870 FTIR Fourier transform IR instrument was
used for IR analysis. Samples were ground into powder and
tableted with potassium bromide before test. The transmission
electron microscopy (TEM) analysis of the catalysts was carried
out on a JEOL 2010 instrument at an accelerating voltage of
200 kV. The samples were ultrasonically dispersed into an ethanol
solution, dropped on the surface of a Cu grid, and dried at RT.
Published by NRC Research Press

Xu et al.
309
Fig. 1. (a) C1s and (b) O1s XPS spectra of the raw and functionalized
Fig. 2. CV curves for electrodes of r-CNTs, f-CNTs-1, and f-CNTs-2 in
CNTs.
1.0 mol/L H2SO4.
Fig. 3. FTIR spectra of (a) r-CNTs, (b) f-CNTs-1, and (c) f-CNTs-2.
Table 1. The atomic percentages of C and O at the
sample surface.
(Fig. 4a). For f-CNTs-1, Pd NPs disperse on the surface of CNT sup-
port with a large particle size distribution (1ϳ8 nm, Fig. 4b). In the
case of f-CNTs-2, Pd NPs uniformly disperse on the surface of CNT
support with a narrow particle size distribution (1ϳ5 nm, Fig. 4c).
Therefore, the dispersion of Pd NPs on the CNT surface relies on
the number of oxygen-containing functional groups, with a larger
number of oxygen-containing functional groups resulting in
higher dispersion of Pd NPs on CNT surface. Similar findings have
been reported by many authors with respect to the relationship
between NP distribution and support modification. For example,
Vu et al. found the dispersion of Pt, Ru, and Pt-Ru alloy NPs on
nano-structured carbon (MWCNTs, SWCNTs, and graphite nano-
fibers) depends on the content of oxygen-containing functional
groups, with a higher content of oxygen-containing functional
groups concomitant with better dispersion of NPs on the surface
of nano-structured carbon.24 Calvo et al. found that the dispersion
of Pd NPs on HNO3-treated activated carbon was higher than that
on an untreated one.25,26 The fact that the surface modification of
carbon materials can improve the dispersion of metal NPs on
them may be associated with the following reasons. First, surface
modification can reduce the hydrophobicity of carbon materials,
which facilitates the even dispersal of metal precursor.27 Second,
the oxygen-containing functional groups on the surface of carbon
H2SO4/HNO3
Samples
treatments
C%
O%
r-CNTs
f-CNTs-1
f-CNTs-2
Pristine
1:1 (v/v), RT
3:1 (v/v), 60 °C
99.324
95.605
91.362
0.676
4.395
8.638
the aromatic ring, the 1562 cm−1 peak in the FTIR spectra can be
attributed to CNTs.23 The peaks at 1715 and 3428 cm−1 are the
characteristic absorption bands of –C=O and –OH, respectively.22
The number of functional groups can be qualitatively determined
by the band intensity. The 1715 and 3428 cm−1 band intensities
were enhanced in the order: r-CNTs < f-CNTs-1 < f-CNTs-2. This
indicates that the number of oxygen-containing functional
groups on the surface of CNTs evolved in the same order, consis-
tent with the CV and XPS characterizations.
Characterization of Pd/CNTs catalysts
Figure 4 shows the TEM images of various Pd/CNT catalysts. For
r-CNTs, the agglomeration of Pd NPs can be clearly detected
Published by NRC Research Press

310
Can. J. Chem. Vol. 91, 2013
Fig. 4. TEM images of (a) Pd/r-CNTs, (b) Pd/f-CNTs-1, and (c) Pd/f-CNTs-2.
materials can prevent the agglomeration of NPs and stabilize
them.28
Fig. 5. XRD patterns of (a) Pd/r-CNTs, (b) Pd/f-CNTs-1, (c) and
Pd/f-CNTs-2.
Figure 5 shows the XRD patterns of Pd/CNT catalysts. The dif-
fraction peaks at 26° correspond to (002) diffraction of graphite
and is characteristic of CNTs.14 For Pd/r-CNTs, the diffraction
peaks at 40.1°, 46.7°, and 68.7° correspond to (111), (200), and
(220) diffraction of metallic Pd, respectively.29 According to the
Scherrer equation, the average grain size of Pd in Pd/r-CNTs cata-
lyst is ϳ36 nm. For Pd/f-CNTs-1, the diffraction peaks of metallic Pd
show a marked decrease in intensity, suggesting higher disper-
sion of Pd NPs on the surface of f-CNTs-1. In the case of Pd/f-CNTs-2,
the diffraction peaks for metallic Pd are essentially not observed,
which is in accordance with the fact that Pd NPs are highly dis-
tributed on the surface of f-CNTs-2.
Figure 6 shows the Pd3d5/2 core level spectra of Pd/CNT catalysts.
For all the samples, the Pd3d5/2 core level peaks can be deconvoluted
into 2 components: the 335.3 eV component corresponding to
Pd(0) and the 337.2 eV component corresponding to Pd(II). It is
noted that the binding energies of Pd(II) remain constant, while
those of Pd(0) shift higher in the order r-CNTs < f-CNTs-1 < f-CNTs-2. It
has been reported that the binding energy of metallic NPs shifts
positively with enhanced NP dispersion.30 Therefore, the XPS re-
sults in Fig. 7 demonstrate that the dispersion of Pd NPs on CNT
surface follows the order Pd/f-CNTs-2 > Pd/f-CNTs-1 > Pd/r-CNTs,
which is consistent with TEM and XRD characterizations.
evolution of bromobenzene conversion as a function of reaction
time over Pd/CNT and Pd/AC catalysts. Regardless of surface mod-
ification, Pd/CNT catalysts show higher activity than Pd/AC cata-
lysts in the hydrodehalogenation of bromobenzene. This is due to
the unique structure and properties of CNTs, which might induce
peculiar activity and selectivity in catalytic reactions.13,31 In the
case of CNT-supported Pd catalysts, consistent with the dispersion
Performance of Pd/CNT catalysts in hydrodehalogenation of
bromobenzene
The hydrodehalogenation of bromobenzene was employed as a
probe reaction to examine the effect of surface modification of
CNTs on the activities of Pd/CNT catalysts. Figure 7 shows the
Published by NRC Research Press

Xu et al.
311
Fig. 6. XPS spectra and curve-fitting of the Pd3d5/2 peak for
(a) Pd/r-CNTs, (b) Pd/f-CNTs-1, and (c) Pd/f-CNTs-2.
Table 2. Effect of solvent on the activity and selectivity of
Pd/f-CNTs-2.
Reaction
time (h)
Conversion
(%)
Selectivity
(%)
Entry
Solvent
1
Methanol
Ethanol
i-Propanol
Octane
THF
2
4
6
6
6
6
100
100
100
100
74
>99.99
>99.99
>99.99
>99.99
>99.99
>99.99
2
3
4
5
6
H2O
91
Note: Reaction conditions: catalyst, 50 mg; bromobenzene, 5 mmol;
hydrazine, 2 mL; solvent, 4 mL; reaction temperature, 60 °C.
Effect of reaction conditions on catalytic performance
Effect of solvent
Recent studies indicated that solvents play an important role
in the reaction of aryl halide dehalogenation with respect to
reaction efficiency and selectivity.6 In our studies, the effects of
solvents on the catalytic performance of Pd/f-CNTs-2 were investi-
gated. Table 2 lists the activities and selectivities of bromoben-
zene hydrodehalogenation over Pd/f-CNTs-2 with various solvents.
It is found that solvents do not exert any effect on reaction selec-
tivity, and for each solvent, benzene was found to be the only
product. In contrast, reaction activity was greatly influenced by
solvents; except for THF, reactions with other organic solvents
demonstrated good bromobenzene conversion with the substrate
almost transformed completely. The catalyst activity decreased
slightly when water was used as the reaction medium, and this
may be attributed to the poor dispersion of the catalyst in water.
These results also remind us that through surface modification of
CNTs to reduce catalyst hydrophobicity and increase their disper-
sion in water, it is possible to achieve high dehalogenation activity
with CNT-supported catalysts in aqueous phase.
Effect of reaction temperature
Figure 8 shows the effects of temperature on the activity and
selectivity of Pd/f-CNTs-2 catalyst in the reaction of bromobenzene
hydrodehalogenation. Lower activity was observed at RT, for
which bromobenzene conversion reached ϳ30% after 6 h of reac-
tion. As the reaction temperature was increased to 40 °C, the
activityofthecatalystwasenhancedsignificantly, withbromoben-
zene conversion up to 94%. As the reaction temperature was in-
creased to >60 °C, bromobenzene could be converted completely
in a short reaction time. However, the decomposition speed of
hydrazine hydrate was accelerated when the reaction temper-
ature exceeded 80 °C. Therefore, the appropriate reaction tem-
perature for bromobenzene hydrodehalogenation reaction
was 60 °C. It is worth noting that benzene selectivities were
always >99.99% throughout the temperature range investi-
gated, indicating that coupling reactions did not occur during
the reaction.
Fig. 7. Conversion of bromobenzene as a function of reaction
time over various catalysts. Reaction conditions: catalyst, 50 mg;
bromobenzene, 5 mmol; hydrazine hydrate (85%), 2 mL; methanol,
8 mL; reaction temperature, 60 °C.
Effect of hydrazine hydrate amount
Cellier et al. demonstrated that the ratio of hydrazine hydrate
to substrate significantly affect the activity and selectivity of cat-
alysts in the dehalogenation of aryl halides.32 Therefore, the ef-
fects of hydrazine hydrate amount on catalyst performance of
Pd/f-CNTs-2 were investigated (Table 3). It was found that a small
amount of hydrazine hydrate can realize the complete conversion
of bromobenzene. An excess of hydrazine hydrate added to reac-
tion system would cause catalytic activity to decline slightly. How-
ever, the benzene selectivity was essentially independent of
hydrazine hydrate amount.
of Pd NPs on CNTs surface, the activities decline in the sequence:
Pd/f-CNTs-2 > Pd/f-CNTs-1 > Pd/r-CNTs. This illustrates the impor-
tant role of CNT modification in determining Pd activity in the
hydrodehalogenation of bromobenzene.
Effect of Pd content
Blank experiments showed that CNTs were inactive to the reac-
tion, indicating that the catalyst active sites are Pd. The effects of
Published by NRC Research Press

312
Can. J. Chem. Vol. 91, 2013
Fig. 8. Effect of reaction temperature on the activity and selectivity
of Pd/f-CNTs-2. Reaction conditions: catalyst, 50 mg; bromobenzene,
5 mmol; hydrazine hydrate (85%), 2 mL; methanol, 4 mL; reaction
time, 6 h.
Fig. 9. Effect of Pd loading on the activity and selectivity of
Pd/f-CNTs-2. Reaction conditions: catalyst, 50 mg; bromobenzene,
5 mmol; hydrazine hydrate (85%), 2 mL; methanol, 4 mL; reaction
temperature, 60 °C; reaction time, 6 h.
Table 3. Effect of N2H4·H2O amount on the activ-
ity and selectivity of Pd/f-CNTs-2.
the adsorption of substrate onto the catalyst.13 Both p-chlorophe-
nol and p-chlorobenzoic acid (Table 4, entries 7 and 8) were com-
pletely converted to phenol and benzoic acid, respectively. For
nitro-halogenated benzene (Table 4, entries 9–11), dehalogenation
reactions occur, concomitant with the reduction of –NO2 due to
the strong reduction of hydrazine hydrate. However, with formic
acid and triethyl ammonium as hydrogen transfer reagents, Cor-
tese et al. found that halogen could be easily removed but –NO2
was intact.33 Heterocyclic aromatic hydrocarbons, such as 3,5-
dibromo pyridine was completely reduced to pyridine (Table 4,
entry 12), indicating the versatility of the Pd/f-CNTs-2 catalyst in
the dehalogenation reaction of aryl halides. It is noteworthy that
for all the substrates studied, biphenyl by-products were not de-
tected. This is probably because the lower reaction temperature
that is unfavorable for by-product formation and the Pd/f-CNTs-2
catalysts that possess high selectivity in the dehalogenation of
aryl halides are employed. In contrast, Busch et al. found a large
number of biphenyl products generated during the dehalogenation
of aryl halides with hydrazine hydrate as hydrogen donor and
Pd/CaCO3 as catalyst.34 Bamfield et al. also found the phenomenon
during the dehalogenation of aryl halides under low temperature
with sodium formate as hydrogen donor and Pd/C as catalyst.35
Hydrazine
amount (ml)
Conversion
(%)
Selectivity
(%)
Entry
1
0.5
1
2
3
4
100
100
100
98
>99.99
>99.99
>99.99
>99.99
>99.99
2
3
4
5
98
Note: Reaction conditions: catalyst, 50 mg; bromoben-
zene, 5 mmol; methanol, 4 mL; reaction temperature,
60 °C; reaction time, 6 h.
Pd content on catalytic efficiency were investigated (Fig. 9). It was
found that the catalytic activity was closely related to Pd content in
the catalyst. When the Pd content was 0.5 wt%, the catalyst was
almost inactive; when Pd content was increased to 1 wt%, the catalyst
showed moderate activity with bromobenzene conversion reaching
51%. As the Pd content was increased to above 2 wt%, the catalyst
showed enhanced activity with the substrate transforming com-
pletely in a short reaction time. Therefore, the Pd content was fixed
at 2 wt% in the following studies to reduce catalyst cost.
Conclusions
Dehalogenation of various aryl halides
Under the optimal reaction conditions discussed for bromoben-
zene hydrodehalogenation, various aryl halides were selected as
substrates to further study the performance of Pd/f-CNTs-2 cata-
lyst in the hydrodehalogenation of aryl halides (Table 4). It was
found that all the substrates selected can be efficiently converted
into the corresponding aromatic hydrocarbons. Because iodine is
a better leaving group compared with chlorine, the activity for
iodobenzene is higher than that for chlorobenzene (Table 4, en-
tries 1 and 2). For the same reason, the activities and selectivities
for dichlorobenzene were lower than those for dibromobenzene
(Table 4, entry 3–6). The position of the –X group was found to
affect the performance of the catalyst with o-dichlorobenzene dis-
playing lower activity and selectivity than p-dichlorobenzene
(Table 4, entries 3 and 4). This may be related to the steric effects
of two neighboring chlorine atoms that influence the adsorption
of the substrate. A similar phenomenon was found by Kawabata
et al., who found that the reaction rate of supported Pd catalysts
for hydrodehalogenation of o-chlorotoluene was lower than that
for p-chlorotoluene, due to the steric effects of methyl that affects
The oxidative functionalization of CNTs by concentrated H2SO4/
HNO3 under ultrasonication has been studied. Two methods were
used for the functionalization of CNTs. One employed a 1:1 (v/v)
mixture of concentrated H2SO4/HNO3, with sonication at RT, the
other employed a 3:1 (v/v) mixture of concentrated H2SO4/HNO3,
with sonication at 60 °C. Compared with raw CNTs, which possess
few surface functional groups, the functionalized CNTs have a
large amount of surface oxide groups, especially in strong oxida-
tion power (H2SO4/HNO3 = 3:1, 60 °C). Deposition of Pd nanopar-
ticles on CNTs was found to be dependent on the amount of
surface oxide groups, as confirmed by TEM and XRD patterns.
Conglomeration of Pd nanoparticles was observed on raw CNTs,
whereas smaller Pd nanoparticles with a narrow size distribution
were obtained over the functionalized CNTs, and the more surface
oxide groups, the better dispersion of Pd nanoparticles on the
surface of CNTs.
The performance of Pd/CNT catalysts in the hydrodehalogenation
of bromobenzene relies on the dispersion of Pd NPs on CNT sur-
faces, with a higher Pd dispersion resulting in higher activity. The
Published by NRC Research Press

Xu et al.
313
Table 4. The hydrodehalogenation of various aryl halides over Pd/f-CNTs-2 catalyst.
Entry
1
Substrate
Product
Conversiona (%)
82
Selectivitya (%)
>99.99
Cl
I
2
100
80
>99.99
90
3b
Cl
Cl
Cl
4b
5
95
92
Cl
100
>99.99
Br
Br
6
7
8
100
100
100
>99.99
>99.99
>99.99
Br
Cl
Cl
Br
OH
OH
O
O
9
100
>99.99
Cl
Br
NO2
NH2
NH2
NH2
10
11
100
100
>99.99
>99.99
NO2
I
NO2
Br
12b
100
>99.99
Br
N
N
Note: Reaction conditions: substrste, 5 mmol; catalyst, 50 mg; hydrazine hydrate (85%), 2 mL; CH3OH, 4 mL;
60 °C; 6 h.
aDetermined by GC.
bSubstrate, 2.5 mmol.
effects of various parameters, such as solvent, reaction tempera-
ture, hydrazine hydrate amount, and Pd loading, on the perfor-
mance of Pd/CNT catalysts in the hydrodehalogenation of
bromobenzene have been investigated systematically. Bromoben-
zene (5 mmol) can be efficiently debrominated at 60 °C over 2 wt%
Pd/CNT catalyst (50 mg) with hydrazine hydrate (0.5–2 mL)
as the hydrogen donor and methanol (4 mL) as the solvent. The
substrates were extended to various aryl halides under the op-
timized reaction conditions, and in all cases Pd/CNT catalysts
show good performance in the hydrodehalogenation of aryl
halides.
Acknowledgements
This work was financially supported by the Doctoral Founda-
tion of Henan University of Technology (No, 2009BS039), the Nat-
ural Science Foundation of the Education Department of Henan
Province (No. 12A150006), and the Scientific and Technological
Project of Zhengzhou (No. 20120724).
References
(1) Morra, M. J.; Borek, V.; Koolpe, J. J. Environ. Qual. 2000, 29 (3), 706. doi:10.
2134/jeq2000.293706x, 10.2134/jeq2000.00472425002900030004x.
(2) Morris, P. J.; Mohn, W. W.; Quenson, J. F.; Tiedje, J. M.; Boyd, S. A. Appl.
Environ. Microbiol. 1992, 58 (9), 3088.
Published by NRC Research Press

314
Can. J. Chem. Vol. 91, 2013
(3) Gampine, A.; Eyman, D. P. J. Catal. 1998, 179 (1), 315. doi:10.1006/jcat.1998.
(20) Martínez, M. T.; Callejas, M. A.; Benito, A. M.; Cochet, M.; Seeger, T.;
Ansón, A.; Schreiber, J.; Gordon, C.; Marhic, C.; Chauvet, O.; Fierro, G. J. L.;
Maser, W. K. Carbon 2003, 41 (12), 2247. doi:10.1016/S0008-6223(03)00250-1.
(21) Datsyuk, V.; Kalyva, M.; Papagelis, K.; Parthenios, J.; Tasis, D.; Siokou, A.;
Kallitsis, I.; Galiotis, C. Carbon 2008, 46 (6), 833. doi:10.1016/j.carbon.2008.
02.012.
(22) Wei, Z. D.; Yan, C.; Tan, Y.; Li, L.; Sun, C. X.; Shao, Z. G.; Shen, P. K.;
Dong, H. W. J. Phys. Chem. C 2008, 112 (7), 2671. doi:10.1021/jp709936p.
(23) Li, X. H.; Niu, J. L.; Zhang, J.; Li, H. L.; Liu, Z. F. J. Phys. Chem. B 2003, 107 (11),
2453. doi:10.1021/jp026887y.
2223.
(4) Huang, Z. Z.; Tang, Y. J. Org. Chem. 2002, 67 (15), 5320. doi:10.1021/jo025693b.
(5) Benyei, A. C.; Lehel, S.; Joo, F. J. Mol. Catal. A 1997, 116 (3), 349. doi:10.1016/
S1381-1169(96)00369-X.
(6) Xia, C.; Xiu, J.; Wu, W.; Liang, X. Catal. Commun. 2004, 5 (8), 383. doi:10.1016/
j.catcom.2004.04.006.
(7) Aramendía, M. A.; Boráu, V.; García, I. M.; Jiménez, C.; Lafont, F.; Marinas, A.;
Marinas, J. M.; Urbano, F. J. J. Catal. 1999, 187 (2), 392. doi:10.1006/jcat.1999.
2632.
(8) Yuan, G.; Keane, M. A. Appl. Catal. B 2004, 52 (4), 301. doi:10.1016/j.apcatb.
2004.04.015.
(9) Nakao, R.; Rhee, H.; Uozumi, Y. Org. Lett. 2005, 7 (1), 163. doi:10.1021/
ol047670k.
(10) Kawabata, T.; Atake, I.; Ohishi, Y.; Shishido, T.; Tian, Y.; Takaki, K.;
Takehira, K. Appl. Catal. B 2006, 66 (3–4), 151. doi:10.1016/j.apcatb.2006.03.
007.
(11) Okumura, K.; Nota, K.; Yoshida, K.; Niwa, M. J. Catal. 2005, 231 (1), 245.
doi:10.1016/j.jcat.2004.12.023.
(24) Vu, H.; Gonçalves, F.; Philippe, R.; Lamouroux, E.; Corrias, M.; Kihn, Y;
Plee, D.; Kalck, P.; Serp, P. J. Catal. 2006, 240 (1), 18. doi:10.1016/j.jcat.2006.
03.003.
(25) Calvo, L.; Gilarranz, M. A.; Casas, J. A.; Mohedano, A. F.; Rodríguez, J. J. Appl.
Catal. B 2006, 67 (1–2), 68. doi:10.1016/j.apcatb.2006.04.016.
(26) Calvo, L.; Gilarranz, M. A.; Casas, J. A.; Mohedano, A. F.; Rodríguez, J. J. Ind.
Eng. Chem. Res. 2005, 44 (17), 6661. doi:10.1021/ie0503040.
(27) Rosca, I. D.; Watari, F.; Uo, M.; Akasaka, T. Carbon 2005, 43 (15), 3124. doi:10.
1016/j.carbon.2005.06.019.
(12) Serp, P.; Corrias, M.; Kalck, P. Appl. Catal. A 2003, 253 (2), 337. doi:10.1016/
S0926-860X(03)00549-0.
(13) Yoon, B.; Wai, C. M. J. Am. Chem. Soc. 2005, 127 (49), 17174. doi:10.1021/
ja055530f.
(28) Albers, P.; Burmeister, R.; Seibold, K.; Prescher, G.; Parker, S. F.; Ross, D. K.
J. Catal. 1999, 181 (1), 145. doi:10.1006/jcat.1998.2279.
(29) Ganesan, M.; Freemantle, R. G.; Obare, S. O. Chem. Mater. 2007, 19 (14), 3464.
doi:10.1021/cm062655q.
(14) Chen, X.; Hou, Y.; Wang, H.; Cao, Y.; He, J. J. Phys. Chem. C 2008, 112 (22), 8172.
doi:10.1021/jp800610q.
(15) Liu, Z. Q.; Ma, J.; Cui, Y. H. Carbon 2008, 46 (6), 890. doi:10.1016/j.carbon.2008.
02.018.
(16) Chen, L.; Yang, K. L.; Liu, H.T; Wang, X. L. Carbon 2008, 46 (15), 2137. doi:10.
1016/j.carbon.2008.09.049.
(17) Kim, J. Y.; Jo, Y.; Lee, S.; Choi, H. C. Tetrahedron Lett. 2009, 50 (46), 6290.
doi:10.1016/j.tetlet.2009.08.121.
(18) Chen, L.; Liu, H. T.; Yang, K. L.; Wang, J. K.; Wang, X. L. Mater. Chem. Phys.
2008, 112 (2), 407. doi:10.1016/j.matchemphys.2008.05.081.
(19) Xing, Y; Li, L.; Chusuei, C. C.; Hull, R. V. Langmuir 2005, 21 (9), 4185. doi:10.
1021/la047268e.
(30) Luo, K.; Lai, X.; Yi, C. W.; Davis, K. A.; Gath, K. K.; Goodman, D. W. J. Phys.
Chem. B 2005, 109 (9), 4064. doi:10.1021/jp045948k.
(31) Planeix, J. M.; Coustel, N.; Coq, B.; Bretons, V.; Kumbhar, P. S.; Dutartre, R.;
Geneste, P.; Bernier, P.; Ajayan, P. M. J. Am. Chem. Soc. 1994, 116 (17), 7935.
doi:10.1021/ja00096a076.
(32) Cellier, P. P.; Spindler, J. F.; Taillefer, M.; Cristau, H. J. Tetrahedron Lett. 2003,
44 (38), 7191. doi:10.1016/S0040-4039(03)01789-1.
(33) Cortese, N. A; Heck, R. F. J. Org. Chem. 1977, 42 (22), 3491. doi:10.1021/
jo00442a008.
(34) Busch, M; Schmidt, W. Chem. Ber. B 1929, 62 (9), 2612. doi:10.1002/cber.
19290620925.
(35) Bamfield, P.; Quan, P. M. Synthesis 1978, (7), 537. doi:10.1055/s-1978-24804.
Published by NRC Research Press