Multiwalled carbon nanotubes oxidized by UV/H2O2 as catalyst for oxidative dehydrogenation of ethylbenzene

Article abstract of DOI:10.1016/j.catcom.2010.11.007

The effect of hydroxyl radical (^*OH)-assisted oxidation of multiwalled carbon nanotubes (MWCNTs) as catalysts for oxidative dehydrogenation of ethylbenzene (ODEB) to styrene was studied. Fourier transform infrared spectroscopy (FTIR), Raman spe

Full text of DOI:10.1016/j.catcom.2010.11.007

Catalysis Communications 12 (2011) 464–469
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Multiwalled carbon nanotubes oxidized by UV/H O as catalyst for oxidative
2
2
dehydrogenation of ethylbenzene
a
a
a
b
c
a,
N.V. Qui , P. Scholz , T. Krech , T.F. Keller , K. Pollok , B. Ondruschka ⁎
a
Institute of Technical Chemistry and Environmental Chemistry, Friedrich-Schiller-University Jena, Lessingstr. 12, 07743 Jena, Germany
Institute of Material Science and Technology, Friedrich-Schiller-University Jena, Löbdergraben 32, 07743 Jena, Germany
Bavarian Research Institute of Experimental Geochemistry and Geophysics, University of Bayreuth, Universitätstr. 30, 95447 Bayreuth, Germany
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 July 2010
Received in revised form 20 October 2010
Accepted 3 November 2010
Available online 12 November 2010
The effect of hydroxyl radical (⁎OH)-assisted oxidation of multiwalled carbon nanotubes (MWCNTs) as
catalysts for oxidative dehydrogenation of ethylbenzene (ODEB) to styrene was studied. Fourier transform
infrared spectroscopy (FTIR), Raman spectroscopy and X-ray photoelectron spectroscopy (XPS) results
2 2
indicated that the presence of hydroxyl radicals during UV/H O oxidative treatment introduced significantly
the amount of hydroxyl and carbonyl groups onto the surface of carbon nanotubes, leading to an
improvement in the catalytic behavior. In the present work, UV/H35 MWCNTs obtained the best catalytic
behavior with 91% styrene selectivity and 47% ethylbenzene conversion at 400 °C under conventional heating.
Keywords:
Oxidative dehydrogenation
Ethylbenzene
2
Transmission electron microscopy (TEM) of UV/H35 MWCNTs showed the formation of new sp -carbon
Multiwalled carbon nanotubes
UV light
layers on the wall of nanotubes after the stability test.
© 2010 Elsevier B.V. All rights reserved.
Hydrogen peroxide
1
. Introduction
More than ninety percent of the styrene monomer used in the
micro pores by coke deposit, and the combustion of the catalyst in air
at high temperature [11,12]. Therefore, the potential application of
the activated carbon in ODEB is not possible. Recently, a new family of
2
world's polymer industry is currently synthesized by the catalytic
dehydrogenation of ethylbenzene. However, this process faces some
disadvantages, including the endothermic reaction requiring lots of
energy, the thermodynamic limitation causing an incomplete reac-
tion, and the deactivation of the catalysts with working time. In order
to overcome these problems, oxidative dehydrogenation of ethylben-
zene (ODEB) has been proposed as a promising alternative [1]:
non-planar sp -hybridized nanocarbons, including nanofilaments,
onion-like carbons and carbon nanotubes (CNTs), have been studied
in ODEB by a few research groups [13–20]. CNTs used as direct catalyst
in ODEB were previously reported by Mestl et al. [14] and Ledoux et al.
[15]. The results of Mestl's group showed that CNTs displayed higher
activity, selectivity and yield than those of graphite. They have also
observed that after catalytic reaction, CNTs have an altered wall
structure: the outer shell of ill-defined carbon layers is combusted and
a thin layer of polymeric carbon deposits covers the outer surface of
the conical graphite edges between the conical layers. Additionally,
the ends of the conical layers seem to be partially oxidized. These
partially oxidized prism faces presumably play an important role in
the catalytic reaction. It was suggested that quinone/hydroxyl groups
presented on the carbon-based catalysts could be the active sites for
oxidative dehydrogenation and a redox mechanism of the Mars van
Krevelen type involving these groups was proposed [12–15], in
agreement with earlier proposals [2–4].
C H C H þ =2O →C H C H þ H O ΔHꢀ ¼ −124:3 kJ mol
1
−
ð1Þ
6
5
2
5
2
6
5
2
3
2
298
A number of catalysts have been tested in ODEB, mainly acidic
oxides [2–4], transition metal oxides [5–7], and carbon-based
materials [8–19]. Most oxide catalysts are active at much higher
temperatures (450–600 °C) than carbon-based catalysts (300–
4
50 °C). One interesting feature of oxide catalysts is that “active
coke” formed on them that has been considered as the real active
catalyst during ODEB [2–4], and thus carbon-based materials are
potential catalysts for ODEB.
2
It has been demonstrated that a pre-treatment of sp -nanocarbon
Activated carbon have been tested successfully in ODEB [8–12]. It
showed a high catalytic activity during the first hours of reaction.
Afterward, its activity is gradually decreased due to the blockage of
can alter their textural features and chemical surface, leading to an
impact on their catalytic performance. Nigrovski et al. [19] proved that
CNTs treated by acid oxidation achieved high catalytic conversion and
selectivity in ODEB. In contrast, CNTs eliminated functional groups by
treatment in an inert gas at high temperature showed a decrease in
the initial catalytic performance, but the regeneration of the
functional groups could be obtained during the reaction [17].
⁎
Corresponding author. Tel.: +49 3641 948400; fax: +49 3641 948402.
E-mail address: bernd.ondruschka@uni-jena.de (B. Ondruschka).
1566-7367/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.11.007

N.V. Qui et al. / Catalysis Communications 12 (2011) 464–469
465
Since their discovery in 1991 by Iijima [21], carbon nanotubes
CNTs) have been regarded as promising new materials because of their
Hydrogen peroxide H
60 wt.% H
(Germany).
2 2
The UV/H O oxidation of MWCNTs was performed in a 1 l glass-
vessel photoreactor. A Hanau low-pressure mercury vapor UV lamp
emitting mainly at 254 nm with an input of 15 W acted as a photon
supplying source. The UV lamp was inserted into a hollow quartz tube
and located at the center of the reactor. The reactor was wrapped
completely with aluminum foil in order to prevent UV exposure.
Typically, 3 g as-received MWCNTs was put into the reactor while the
2
O
2
with various concentrations (35, 50, and
(
2 2
O aqueous solutions) was purchased from Solvay GmbH
unique electrical and mechanical properties. Most of CNT applications
require functionalization, such as improving their solubility in different
media, attaching different groups or even metallic particles for further
utilization. The oxidative treatment of CNTs has been recognized as an
efficient method for purification, promoting dispersion, and surface
activation [22]. Typically, it generates various kinds of surface
functional groups such as carboxylic (–COOH), carbonyl (–C=O) and
hydroxyl groups (–C–OH). The oxidative treatment has been
commonly carried out under severe conditions, e.g., in concentrated
2 2
UV light was turned on, then 500 ml of H O solution at various
hot nitric acid or in mixtures of H
generation of oxygenated functional groups, oxidative treatments
can also damage the structure of CNTs [24,25].
2
SO
4
/KMnO
4
[22]. Besides the
concentrations was added slowly under vigorous stirring (250 rpm).
The oxidative treatment of MWCNTs was carried out at 15 °C for 7.5 h.
Next, the reaction mixtures were filtered through the wide pore filter
paper. Then the black solid was washed 3 times with de-ionized water
and dried overnight in an oven at 120 °C. Finally, oxidized MWCNTs
were stored in the desiccator after grinding and sieving at a fraction of
2 2
Hydrogen peroxide (H O ) is one of the most powerful oxidants
and an environmentally friendly chemical since no harmful by-
products remain [26]. It could be used in the chemical oxidation of
CNTs, such as utilizing alone [27] or in combination with acids
250 μm. A sample of MWCNTs oxidized with 35 wt.% H
the dark (without UV lamp) was also prepared. The samples were
named as H35 MWCNTs oxidized with 35 wt.% H solution.
UV/H35, UV/H50 and UV/H60 MWCNTs oxidized with 35, 50 and
60 wt.% H solution in combination with UV irradiation, respectively.
2 2
O solution in
[
22,27,29]. Kuznetsova et al. [28] reported that both carbonyl and
ether groups could be introduced onto surface of single-walled carbon
nanotubes (SWNTs) during the oxidation with H SO /H . The
results of Rasheed et al. [23] indicated that H SO /H for the
treatment of carbon nanofibers (CNFs) created at least defects and
acidic sites among the employed oxidizing agents, including HNO
SO /HNO , RuO /NaOCl, KMnO and H SO /H . Moreover, Aviles
et al. [29] have reported that H promoted an increase in hydroxyl
2 2
O
2
4
2 2
O
2
4
2
O
2
2 2
O
Fourier transform infrared spectroscopy (FTIR) of samples was
measured with KBr at room temperature on a Bruker IFS 66
spectrometer. Each spectra was smoothed (Savitsky–Golay-5 point
function) and baseline corrected. Thermogravimetric analyses (TGA)
were performed on a TGA-60 instrument (Shimadzu) under nitrogen
atmosphere from room temperature to 750 °C with heating rate of
3
,
H
2
4
3
2
4
2
4
2 2
O
2 2
O
and carbonyl groups on the surface of MWCNTs.
Hydrogen peroxide is also employed in the advanced oxidation
−1
processes (AOPs), such as ozone/H
2
2
O , UV/H
2
O
2
2
, UV/H O
2
/ozone,
20 K min . Transmission electron microscopy (TEM) was done at
200 keV on a LEO 922 instrument. The sample was first dispersed in
with soft grinding in a mortar and then a drop of suspension put on
Cu-Lacey-grids. Raman spectra of samples were measured in a range
Fenton and Photo-Fenton oxidation [30]. The AOPs dealing with
highly potent chemical oxidants such as OH have been used for the
oxidation and destruction of organic contaminants in pesticides [31]
⁎
−1
or in dye effluents [32]. Recently, it has been found that highly
500–2000 cm
at an excitation wavelength of 532 nm on a JY-HR
⁎
⁎
⁎
oxidative species, such as O , H atoms and OH radicals, could also be
800 Raman spectrometer, equipped with a Nd:YAG laser source. X-ray
photoelectron spectroscopy (XPS) of samples was measured at room
temperature on a Quantum 2000 instrument with a focused
monochromatic Al Kα source at 1486.6 eV for excitation.
utilized as oxidizing agents for the oxidation of CNTs [33–35]. Imasaka
⁎
⁎
et al. [33] conducted O and H atoms generated in water by a pulsed
streamer discharge in order to oxidize MWCNTs. Based on the surface
analysis results of CNTs after the treatment with Fenton reagents, Li et
⁎
al. [35] suggested that OH radicals with double properties of
2.2. ODEB catalytic testing
oxidizability and electrophilicity enable to react with C=C double
bonds of CNTs to form functional surface groups. Nevertheless,
insoluble iron-sludge (Fe3 ) that is derived from Fenton process
could be encountered in the purification of oxidized nanotubes after
treatment.
Catalytic testing was performed under conventional heating in a
fixed-bed quartz-tube reactor (20 mm i.d., 330 mm length). Typically
1.0 g catalyst was used. A K-type thermocouple was inserted into the
catalyst bed for recording the reaction temperature. All stainless steel
transfer lines were kept at 110 °C by heating tape in order to prevent
the condensation of ethylbenzene (EB) and styrene (ST).
+
It is well known that ⁎OH radicals can also be generated by the
direct UV photolysis of hydrogen peroxide:
A flow of nitrogen was used for carrying EB evaporation at 38 °C
(EB partial pressure of 2.584 kPa at 38 °C). Then EB evaporation was
mixed with air. The flow rates of nitrogen and air were controlled by
mass flow controllers (Brooks model 5850) in order to obtain the
desired EB concentration of 2.3 vol.%, EB to oxygen ratio of 1:1, and the
total flow rate was 55 ml min . The inlet and outlet gases were
analyzed with an online gas-chromatograph (Agilent 7890 A),
equipped with two columns, a HP-5 (0.32 mm i.d., 50 m length,
⁎
H O þ hν→2 OH
ð2Þ
2
2
Therefore, UV/H
2
O
2
oxidation of carbon nanotubes would be
considered as an alternative feasibility. Here, we report an efficient
process using UV/H for oxidative treatment of MWCNTs. The
relationship of the chemical surface of MWCNTs after processing to
their catalytic behavior in ODEB is also studied.
−1
2
O
2
1
5
.05 μm film thickness) for analyzing of hydrocarbons and a molsieve
A PLOT (0.32 mm i.d., 30 m length, 30 μm film thickness) for CO and
2
. Experimental
2
CO , coupled with FID and TCD detectors, respectively. The reactions
were considered in steady state when at least three consecutive
measurements showed constant composition (within 5% deviation).
2
.1. Oxidation of MWCNTs
MWCNTs “Baytube®” used in this study were purchased from
3. Results and discussion
Bayer Technology Services Ltd. According to the product specifica-
tions, MWCNTs were fabricated by chemical vapor deposition
method. They have more than 95% of purity, outside diameter of
3.1. Characterization
1
3–16 nm, inside diameter of 4 nm, and a typical length of 1–10 μm.
The FTIR spectra of as-received, H35 and UV/H35 MWCNTs are
presented in Fig. 1. For as-received MWCNTs, peaks between 2800
MWCNTs were used as-received.

466
N.V. Qui et al. / Catalysis Communications 12 (2011) 464–469
Table 1
(
a)
b)
c)
D- and G-band frequencies and I
MWCNTs.
D G
/I of Raman spectra of as-received, H35 and UV/H35
(
Samples
D-band frequency
G-band frequency
D G
I /I
(
−1
−1
(
cm
)
(cm
)
1742
2855
as-received MWCNTs
H35 MWCNTs
UV/H35 MWCNTs
1348
1348
1348
1583
1583
1592
1.29
1.34
1.42
2922
1059
1630
3434
functional groups could be attached on the surface of MWCNTs after
the oxidative treatments [37]. In addition, the higher relative intensity
4000
3500
3000
2500
2000
1500
1000
I
/I
ratio of UV/H35 MWCNTs in comparison to H35 MWCNTs can be
D
G
-
1
⁎
Wavenumber (cm )
believed that OH radicals generated via UV/H
2
O
2
process strength-
ened the formation of functional surface groups in the oxidative
treatment.
Fig. 1. FTIR spectra of (a) as-received, (b) H35 and (c) UV/H35 MWCNTs in KBr.
The TEM pictures of as-received and UV/H35 MWCNTs are
illustrated in Fig. 3. It can be seen that as-received MWCNTs were
closed at the ends of the tubes. They also contained metallic particles
and amorphous carbon impurities. These impurities were removed
effectively after UV/H O treatment.
and 3500 cm− are attributed to the stretching of C–H. and O–H
1
bonds, which are related to hydroxyl groups. The weak peak at
−1
1
059 cm
indicates the C–O vibration. This result indicated that
oxygenated functional groups presented on as-received MWCNTs due
to an unintentional generation during the manufacturing [29]. After
the treatment with H O and UV/H O , the wavenumbers of oxidized
2 2 2 2
MWCNTs are the same as those of as-received MWCNTs but show
higher absorbance. This increase in the case of C=C stretching at
2
2
HR-TEM pictures of UV/H35 MWCNTs before and after 24 h on
stream under conventional heating are illustrated in Fig. 4. They
showed an increase in the number of walls of UV/H35 MWCNTs. It is
seen that most of the new layers covered were the ordered graphene
2
layers. Nigrovski et al. [19] reported that new sp -carbon layers could
−1
1
630 cm could be explained by the breaking of symmetry of CNTs,
only be formed on the oxidized MWCNTs under microwave heating.
Interestingly, in our study, new graphite layers deposited on the
surface of UV/H35 MWCNTs under conventional heating also.
which enhances the generation of induced electric dipole [41].
Moreover, the O–H and C–O absorbance peaks on UV/H35 MWCNTs
which are higher than those on H35 MWCNTs could be attributed to
the presence of hydroxyl radicals.
Raman spectroscopy was used as a tool for reflecting the structural
CNTs. The Raman spectra of as-received, H35 and UV/H35 MWCNT
samples are shown in Fig. 2.
Each of them consists of two characteristic bands, namely the
−1
D-band at ~1348 and the G-band at ~1583 cm . The D-band is
3
usually attributed to the disorder or sp -hybridized carbon and the G-
band corresponds to the graphite in-plane vibration [35]. It can be
seen that there is no change of frequency of D- and G-bands of H35
MWCNTs. However, the frequency of the G-band of UV/H35 MWCNTs
−1
shifted from 1583 to 1592 cm . This movement could be related to a
significant change of chemical surface of the nanotubes. The intensity
ratio of D- and G-bands (I
D G
/I ) could be used to evaluate this change
[
33–35]. The I /I ratio (Table 1) of the oxidized MWCNTs is higher
D G
than that of as-received MWCNTs, indicating that more oxygenated
D-band
G-band
(a)
(b)
(c)
500
800
1100
1400
1700
2000
-
1
Raman shift (cm )
Fig. 2. Raman spectra of as-received (a), H35 (b), and UV/H35 MWCNTs (c).
Fig. 3. TEM pictures of as-received (A) and UV/H35 MWCNTs (B).

N.V. Qui et al. / Catalysis Communications 12 (2011) 464–469
467
oxidative treatment processes to the chemical surface on MWCNTs is
provided by XPS analysis (Fig. 5). The XPS C1s spectra of as-received,
oxidized MWCNTs before and after ODEB (Fig. 6) can be fitted by
convoluting four peaks, each representing a separate carbon bond,
namely, –C=C (284.6 eV), –C–OH (286.1 eV), –C=O (288 eV) and
–COOH (289.7). Table 2 summarizes the XPS C1s percentage of the
specific peaks of as-received, oxidized MWCNTs before and after
ODEB. As shown in Table 2, functional groups existed on the surface of
as-received MWCNTs. After the oxidative treatments, oxidized
MWCNTs were introduced more functional groups. This fact is
consistent with the results of XPS C1s and FTIR analyses previously.
It is seen that the oxidation of MWCNTs can be assisted by the
presence of hydroxyl radicals in the process. Indeed, at the same
concentration of H O solution (35 wt.%) with UV irradiation, more
2 2
functional groups (hydroxyl and carbonyl groups) are generated on
the surface of CNTs than that without UV irradiation.
2 2 2 2
At higher concentrations than 35 wt.% H O used in UV/H O
process, it is seen that the fraction of C=O and C–OH on UV/H50 and
UV/H60 MWCNT decreased continuously in comparison to UV/H35
MWCNTs, indicating that the available amount of ⁎OH radicals for the
⁎
oxidation possibly can be reduced due to the scavenging of OH
radicals, as expressed by the following equations [39,40]:
⁎
⁎
2
OH þ H O →H O þ HO
ð3Þ
2
2
2
⁎
⁎
HO þ OH→H O þ O
ð4Þ
ð5Þ
2
2
2
⁎
⁎
OH þ OH→H O
2
2
The scavenging decreases the number of ⁎OH radicals used for the
oxidation of CNTs. As a result, it affects negatively to the oxidative
treatment.
Fig. 4. HR-TEM pictures of UV/H35 MWCNTs before (A) and after (B) 24 h time on
stream under thermal heating.
The change in the percentage of functional groups of catalysts
before and after ODEB can be compared (Table 2). The content of
oxygenated groups of both H35 and UV/H35 MWCNTs enhanced after
ODEB. In particular, UV/H35 MWCNTs were mainly added hydroxyl
and carbonyl groups, whereas only more hydroxyl groups on H35
XPS is typically used to analyze the nature of chemical bondings on
the surface of materials. It can provide useful information about
functional groups and defects of CNTs [38]. The influence of the
A
B
282
284
286
288
290
292
282
284
286
288
290
292
Binding energy (eV)
Binding energy (eV)
C
D
282
284
286
288
290
292
282
284
286
288
290
292
Binding energy (eV)
Binding energy (eV)
Fig. 5. XPS C1s spectra of (A) H35, (B) UV/H35, (C) UV/H50 and (D) UV/H60 MWCNTs.

468
N.V. Qui et al. / Catalysis Communications 12 (2011) 464–469
A
282
284
286
288
290
292
Binding energy (eV)
B
Fig. 7. TGA of as-received and oxidized MWCNTs under N
2
flow rate of 30 ml min− and
1
a heating rate of 20 K min-
1
.
styrene selectivity is 88 and 94% at 350 °C, 86 and 92% at 375 °C, 83
and 91% at 400 °C over H35 and UV/H35 MWCNTs, respectively.
Therefore, it is concluded that the UV/H O oxidative treatment
2 2
282
284
286
288
290
292
Binding energy (eV)
remarkably improved the styrene selectivity of the catalyst. Over
UV/H50 and UV/H60 MWCNTs, ethylbenzene conversion and styrene
selectivity increased simultaneously with the rise of reaction
temperature from 350 to 375 °C. They obtained a higher ethylbenzene
conversion but lower styrene selectivity at 400 °C in comparison to
those at 375 °C. It is noted that the higher hydrogen peroxide
Fig. 6. XPS C1s spectra of (A) H35 and (B) UV/H35 MWCNTs after ODEB.
MWCNTs. Moreover, the increase in functional groups of UV/H35
MWCNTs is significantly higher than H35 MWCNTs.
TGA results of as-received and oxidized MWCNTs are presented in
Fig. 7. The weight loss of all the samples can be divided into three
levels, corresponding to the three zones of temperature. Below 150 °C,
all samples lost a minor weight (b1%) corresponding to an
evaporation of the adsorbed water. Between 150 and 500 °C, the
observed weight loss is relatively high, and is attributed to the
decarboxylation of carboxylic groups and the elimination of hydroxyl
groups attached to the sidewall of MWCNTs [29]. The drastic loss
occurring above 500 °C for as-received and above 550 °C for oxidized
MWCNTs could be related to the degradation of disordered or
amorphous carbon [29]. The weight loss of oxidized MWCNTs was
not as strong as as-received MWCNTs. This result indicates that the
oxidative treatments purified the CNTs and eliminated amorphous
carbon substantially, as reported in previous investigations [29,36].
2 2
concentration in the UV/H O oxidative treatment from 35 to 60 wt.%
results in a decrease in of styrene yield from 42 to 35%, respectively.
It is seen that there is a correlation between carbonyl and hydroxyl
groups presented on catalysts and their catalytic activity. Indeed, as
compared the fraction of hydroxyl and carbonyl groups of H35,
UV/H35, UV/H50, UV/H60 MWCNTs and their activity, showed that a
higher hydroxyl and carbonyl surface species result in a better
catalytic performance. So that these surface species might be the
major sites of the reaction.
In order to investigate the stability of the catalysts, long-term
catalytic tests were carried out at 400 °C under conventional heating
for 24 h time on stream (Fig. 8). On the whole, all catalysts displayed a
stable performance. UV/H35 MWCNTs exhibited a slight decrease in
conversion at the beginning of the reaction. After that, its activity
tends toward a slow rising with time. The selectivity to styrene on
UVH35 MWCNTs still maintains the highest value of N90%. Su et al.
[16,17] have suggested that the catalytic behavior of the different
carbon materials has a correlation with the formation of functional
surface groups during the oxidative pre-treatment or during the
induction period. Therefore, the highest activity of UV/H35 MWCNTs
could be possibly linked with the highest content of hydroxyl and
carbonyl groups that were generated not only through the oxidative
3
.2. Catalytic test in ODEB
Table 3 presents the results of ethylbenzene conversion (CEB),
styrene selectivity (SST) and styrene yield (YST) as a function of the
reaction temperatures over the various MWCNT catalysts. Over H35
and UV/H35 MWCNTs, ethylbenzene conversion increased rapidly
while styrene selectivity decreased slightly in the range temperature
from 350 to 400 °C. They obtained the same ethylbenzene conversion
at the same temperature steps but styrene selectivity on UV/H35
MWCNTs still maintains higher than that on H35 MWCNTs. Indeed,
Table 3
The conversion (CEB), selectivity (SST) and yield (YST) styrene versus the reaction
Table 2
temperatures during ODEB on MWCNTs (testing conditions: catalyst weight 1.0 g,
XPS C1s percentage of the specific peaks of as-received, oxidized MWCNTs and oxidized
MWCNTs after ODEB testing.
-1
EB/O
2
/N
2
=2.3:2.3:95.4 (vol.%), total flow 55 ml min ).
Samples
350 °C
375 °C
400 °C
Samples
C=C (%)
C–OH (%)
C=O (%)
COOH (%)
C
EB
S
(%)
ST
Y
(%)
ST
C
(%)
EB
S
(%)
ST
Y
(%)
ST
C
(%)
EB
S
(%)
ST
Y
ST
(%)
as-received MWCNTs
H35 MWCNTs
H35 MWCNTs after ODEB
UV/H35 MWCNTs
UV/H35 MWCNTs after ODEB
UV/H50 MWCNTs
UV/H60 MWCNTs
71.3
71.2
67.0
58.2
44.9
62.5
68.8
16.6
14.4
20.1
26.2
35.8
22.8
17.8
6.9
7.8
7.1
9.0
11.7
8.1
5.3
6.5
5.9
6.6
7.5
6.6
5.9
(%)
as-received MWCNTs
H35 MWCNTs
UV/H35 MWCNTs
UV/H50 MWCNTs
UV/H60 MWCNTs
27
20
20
19
24
65
88
94
81
82
18
18
19
15
20
35
32
32
28
33
64
86
92
89
87
22
28
29
25
29
34
47
46
47
43
61
83
91
80
81
21
39
42
38
35
7.6

N.V. Qui et al. / Catalysis Communications 12 (2011) 464–469
469
Acknowledgments
We wish to thank Dr. Nicolae Tarcea, Institute for Physical
Chemistry, FSU, for conducting Raman spectroscopy and Ralf Wagner,
Institute of Material Science and Technology (IMT), FSU, for his help in
XPS measurement.
References
[
[
1] F. Cavani, F. Trifiro, Appl. Catal. A 133 (1995) 219.
2] T.G. Alkhazov, A.E. Lisovskii, M.G. Safarov, A.M. Dadasheva, Kinet. Katal. 13 (1972)
509.
[
[
[
[
3] Y. Iwasawa, H. Nobe, S. Ogasawara, J. Catal. 31 (1973) 444.
4] G. Emig, H. Hofmann, J. Catal. 84 (1983) 15.
5] J. Hanuza, B.J. Trzebiatowska, W. Oganowski, J. Mol. Catal. 29 (1985) 109.
6] W. Oganowski, J. Hanuza, H. Drulis, W. Mista, L. Makalik, Appl. Catal. A 136 (1996)
1
43.
7] V.M. Zhyznevskiy, R.D. Tsybukh, V.V. Gumenetskiy, V.V. Kochubeiy, Appl. Catal. A
38 (2003) 19.
[
2
[
[
8] G.C. Grunewald, R.S. Drago, J. Mol. Catal. 58 (1990) 227.
9] R.S. Drago, K. Jurczyk, App. Catal. A 26 (1994) 117.
[
[
[
[
[
10] M.F.R. Pereira, J.J.M. Orfao, J.L. Figueiredo, Appl. Catal. A 184 (1999) 153.
11] M.F.R. Pereira, J.J.M. Orfao, J.L. Figueiredo, Appl. Catal. A 218 (2001) 307.
12] M.F.R. Pereira, J.J.M. Orfao, J.L. Figueiredo, Appl. Catal. A 196 (2000) 43.
13] N. Maksimova, G. Mestl, R. Schlögl, Stud. Surf. Sci. Catal. 133 (2001) 383.
14] G. Mestl, N. Maksimova, N. Keller, V. Roddatis, R. Schlögl, Angew. Chem. Int. Ed. 40
(2001) 2066.
[
[
15] M. Ledoux, R. Vieira, C. Pham-Huu, N. Keller, J. Catal. 216 (2003) 333.
16] D.S. Su, N. Maksimova, G. Mestl, V. Kuznetsov, V. Keller, R. Schlögl, N. Keller,
Carbon 45 (2007) 2145.
[
[
[
[
17] D.S. Su, N. Maksimova, J.J. Delgado, N. Keller, G. Mestl, M.J. Ledoux, R. Schlögl,
Catal. Today 102–103 (2005) 110.
18] J.J. Delgado, X. Chen, J.P. Tessonnier, M.E. Schuster, E. Del Rio, R. Schlögl, D.S. Su,
Catal. Today 150 (2010) 49.
19] B. Nigrovski, P. Scholz, K. Krech, N.V. Qui, K. Pollok, T. Keller, B. Ondruschka, Catal.
Commun. 10 (2009) 1473.
20] M.F.R. Pereira, J.L. Figueiredo, J.J.M. Orfao, P. Serp, P. Kalck, Y. Kihn, Carbon 42
(
2004) 2807.
Fig. 8. EB conversion (A) and ST selectivity (B) as a function of time on stream at 400 °C.
Testing conditions: catalyst weight 1.0 g, EB/O /N =2.3:2.3:95.4 (vol.%), total flow
[
[
[
[
[
[
[
21] S. Iijima, Nature 354 (1991) 56.
2
2
22] L. Meng, C. Fu, Q. Lu, Progr. Nat. Sci. 19 (2009) 801.
23] A. Rasheed, J.Y. Howe, M.D. Dadmun, P.F. Britt, Carbon 45 (2007) 1072.
24] P.X. Hou, C. Liu, H.M. Cheng, Carbon 46 (2008) 2003.
25] Y. Miyata, Y. Maniwa, J.H. Kataura, J. Phys. Chem. B 110 (2006) 25.
26] Z. Qiang, J.H. Chang, C.P. Huang, Wat. Res. 36 (2002) 85.
27] Y. Peng, H. Liu, Ind. Eng. Chem. Res. 45 (2006) 6483.
-
1
55 ml min .
treatment but also during the induction period of ODEB. It seems that
the couple of hydroxyl/carbonyl surface groups plays the important
rule of the active sites for this reaction.
[28] A. Kuznetsova, I. Popova, J.T.J. Yates, M.J. Bronikowski, C.B. Huffman, J. Liu, R.E.
Smalley, H.H. Hwu, J.G. Chen, J. Am. Chem. Soc. 123 (2001) 10699.
[
29] F. Aviles, J.V. Caauich-Rodriguez, L. Moo-Tah, A. May-Pat, R. Vargas-Coronado,
Carbon 47 (2009) 2970.
4
. Conclusions
[
[
[
[
30] A. Goi, M. Trapido, Proc. Estonian Acad. Sci. Chem. 50 (2001) 5.
31] W.Z. Tang, C.P. Huang, Environ. Technol. 17 (1996) 1371.
32] M.S. Lucas, J.A. Peres, Dyes Pigm. 71 (2006) 236.
2 2 2 2
The oxidative treatment of MWCNTs with H O and with UV/H O
was applied. The results from the characteristic analysis indicated that
33] K. Imasaka, J. Suehiro, Y. Kanatabe, Y. Kato, Nanotechnology 17 (2006) 3421.
⁎
the presence of OH radicals during the UV irradiation of H
generated more quantity of functional groups than those with H
2
O
O
2
[34] X. Zhang, L. Lei, B. Xia, Y. Zhang, J. Fu, Electrochim. Acta 54 (2009) 2810.
[
35] W. Li, Y. Bai, Y. Zhang, M. Sun, R. Cheng, X. Xu, Y. Chen, Y. Mo, Synth. Met. 155
2005) 509.
2
2
(
treatment only. The XPS C1s results of UV/H50 and UV/H60 MWCNTs
showed that the use of H at concentrations higher than 35 wt.% in
UV/H process did not enhance the content of the functional
[
36] V. Datsyuk, M. Kalyva, K. Papagelis, J. Parthenios, D. Tasis, A. Siokou, I. Kallitsis, C.
Galiotis, Carbon 46 (2008) 833.
2 2
O
2
O
2
[37] S.H. Liao, C.Yu. Yen, C.H. Hung, C.C. Wenig, M.C. Tsai, Y.F. Lin, C.C.M. Ma, C. Pan, A.
Su, J. Mater. Chem. 18 (2008) 3993.
[
[
groups, as expected. This fact could be attributed to the scavenging of
OH radicals. Among the sample prepared, UV/H35 MWCNTs
38] G.S. Wang, S.T. Hsieh, C.S. Hong, Wat. Res. 34 (2000) 3882.
39] C.L. Hsueh, Y.H. Huang, C.C. Chen, J. Mol. Catal. A 245 (2006) 78.
⁎
obtained the highest styrene selectivity of 91% and ethylbenzene
conversion of 47% at 400 °C. New sp -carbon layers on the surface of
UV/H35 MWCNTs were observed after catalytic testing under
conventional heating.
[40] G.S. Wang, S.T. Hsieh, C.S. Hong, Wat. Res. 34 (2000) 3882.
[41] A.G. Osorio, I.C.L. Silveira, V.L. Bueno, C.P. Bergmann, Appl. Surf. Sci. 255 (2008)
2
2485.