Evolution and reactivity of active oxygen species on sp2@sp 3 core-shell carbon for the oxidative dehydrogenation reaction

Article abstract of DOI:10.1002/cctc.201402097

Different sp²@sp³ core-shell structures are obtained on nanodiamond by using annealing treatment at increasingly higher temperatures. The resulting nanocarbons can serve as model catalysts to investigate the structural effect on the evolution and chemical nature of oxygen functional groups for oxidative dehydrogenation reactions. We studied in situ reactions and characterization data and found that the initial existence of oxygen-containing groups on a catalyst surface had a low contribution to the catalytic performance. The active oxygen species can be generated promptly in situ by the chemisorption of O₂ under the reaction conditions and involved in catalytic dehydrogenation process following a redox mechanism. For different hybridized nanostructures, the same types of generated active oxygen groups show different catalytic capabilities, which can be regulated by the sp ²-hybridized carbon fraction of nanodiamond. The ketonic carbonyl groups formed on graphitic onion-like carbon surface are more active and can improve the selectivity to alkenes significantly compared with the initial nanodiamond and traditional carbon nanotubes.

Full text of DOI:10.1002/cctc.201402097

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201402097
Evolution and Reactivity of Active Oxygen Species on
sp2@sp3 Core–Shell Carbon for the Oxidative
Dehydrogenation Reaction
Xiaoyan Sun,[a] Rui Wang,[b] Bingsen Zhang,[a] Rui Huang,[a] Xing Huang,[c] Dang Sheng Su,*[a]
Tong Chen,[d] Changxi Miao,[d] and Weimin Yang[d]
Different sp2@sp3 core–shell structures are obtained on nano-
diamond by using annealing treatment at increasingly higher
temperatures. The resulting nanocarbons can serve as model
catalysts to investigate the structural effect on the evolution
and chemical nature of oxygen functional groups for oxidative
dehydrogenation reactions. We studied in situ reactions and
characterization data and found that the initial existence of
oxygen-containing groups on a catalyst surface had a low con-
tribution to the catalytic performance. The active oxygen spe-
cies can be generated promptly in situ by the chemisorption
of O2 under the reaction conditions and involved in catalytic
dehydrogenation process following a redox mechanism. For
different hybridized nanostructures, the same types of generat-
ed active oxygen groups show different catalytic capabilities,
which can be regulated by the sp2-hybridized carbon fraction
of nanodiamond. The ketonic carbonyl groups formed on
graphitic onion-like carbon surface are more active and can im-
prove the selectivity to alkenes significantly compared with
the initial nanodiamond and traditional carbon nanotubes.
Introduction
Metal-free carbon-based catalysts with sp2- or sp3-hybridized
nanostructures are attractive alternatives to conventional
metal-based catalysts in many industrial processes, such as the
oxidation and hydration of hydrocarbons,[1] dehydrogenation
of alkanes,[2] and electrocatalytic oxygen reduction reactions.[3]
The defect/edge sites of the carbon structure can be modified
by heteroatoms, such as O, N, P, B, or S, which exist on the
carbon surface in the form of various functional groups and
provide a strong redox activity.[4] Many studies on carbon catal-
ysis have suggested that the surface functional groups play an
essential role in a variety of catalytic processes.[5] For example,
surface ketonic carbonyl groups have been confirmed as the
active sites for the oxidative dehydrogenation (ODH) of hydro-
carbons, which is assumed to proceed by a redox mecha-
nism.[6] For various oxygen functional groups on the carbon
surface, the thermally unstable oxygen species may desorb or
decompose under the reaction conditions, whereas the rela-
tively stable oxygen species may be retained or newly formed
on the defects by interacting with oxygen molecules.[7] Howev-
er, few in situ characterization studies have been performed
and used to clarify the dynamic evolutions of oxygen groups
during the catalytic process. Moreover, different hybridizations
of nanocarbon can produce distinct electronic structures and
completely different bulk properties, which will have a great
impact on the evolution and chemical nature of the anchored
oxygen functionalities. To the best of our knowledge, there are
few reports on this structural contribution. Consequently, a de-
tailed study on the evolution and chemical activity of oxygen
species on different nanostructures is necessary to understand
the ODH reaction, which can provide a better understanding
of the catalytic nature of active sites and open possibilities for
the controlled modification of nanocarbon catalysts.
[a] X. Sun, Dr. B. Zhang, R. Huang, Prof. D. S. Su
Shenyang National Laboratory for Materials Science
Institute of Metal Research, Chinese Academy of Sciences
72 Wenhua Road, Shenyang 110016 (China)
Fax: (+86)24-83970019
An appropriate model catalyst is required to vary the hybrid-
ized nanostructures and the nature of the corresponding func-
tional groups. Nanodiamond (ND) is a nanostructure of sp3-
bonded carbon synthesized by the detonation method.[8] The
high surface-to-bulk ratio and unsaturated surface bonding
result in an amorphous carbon that covers the diamond core.[9]
ND is thermodynamically unstable and can be transformed to
sp2-hybridized carbon onions by irradiation or thermal treat-
ment. During the transformation, an intermediate phase of
sp2@sp3 core–shell structures that feature a diamond core en-
cased in an onion-like shell can be formed.[10] These nanocar-
bons that combine controllable sp2/sp3 ratios are excellent
model catalysts to investigate structural contributions for the
evolution and chemical activity of oxygen groups during cata-
E-mail: dssu@imr.ac.cn
[b] Dr. R. Wang
National Institute of Clean-and-Low-Carbon Energy
Changping District, Beijing 102209 (China)
[c] X. Huang
Fritz Haber Institute of the Max Planck Society
Faradayweg 4–6, Berlin 14195 (Germany)
[d] T. Chen, C. Miao, W. Yang
Sinopec Shanghai Research Institute of Petrochemical Technology
1658 Pudong Beilu, Shanghai, 201208 (China)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201402097.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2270 – 2275 2270

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
lytic reactions. Herein, we controlled the formation of the
graphitic shells of ND and modulated the sp2/sp3 ratios by an-
nealing treatment.[11] Therefore, the chemical nature of the oxy-
genated functionalities can be varied, and the relationship be-
tween the evolution of active oxygen species in the ODH reac-
tion and the corresponding catalytic performance is explored.
Results and Discussion
In our previous study,[11] we demonstrated that an optimal cat-
alytic performance for the direct dehydrogenation of propane
(DH) can be obtained on sp2@sp3 core–shell nanocarbon by
regulating the microstructure and surface chemistry of ND. The
high catalytic activity was derived from a high concentration
of structural defects, including vacancies, edges, and adatoms,
which can play the role of active sites under anaerobic reaction
conditions. In this work, we focus on the evolution of oxygen-
containing groups in the ODH reaction as well as the structural
impact on the formation of active oxygen species. To obtain
different sp2/sp3 ratios, ND was annealed at 1000, 1300, and
15008C for 4 h under an inert atmosphere (samples are denot-
ed as ND-T, in which T is the annealing temperature). Based on
our previous results (summarized in Table 1) and high-resolu-
tion transmission electron microscopy (HRTEM) characteriza-
tion (Figure 1), we found that fresh ND had the typical struc-
ture of a crystalline diamond particle covered by a thin and
Figure 1. HRTEM images of a) ND, b) ND-1000, c) ND-1300, and d) ND-1500.
BET surface areas and pore volumes increased significantly
with the increasing annealing temperature, which resulted
from the conversion of the sp3-hybridized diamond
framework into the sp2-bonding of onion-like
carbon.[13]
Table 1. Properties (BET surface area, pore volume, sp2 carbon fraction, ID/IG, O/H con-
tent) of ND-derived catalysts.[11]
To investigate the structure–activity correlation, the
ODH of propane was performed as a probe reaction.
Propene and COx were the main products, and the
catalytic performances during 8 h time on stream
(TOS) are shown in Figure 2. For the fresh ND cata-
lyst, a clear decrease in propane conversion can be
observed in the initial 4 h of the reaction (Figure 2a),
which was likely related to the removal of active sites
present on the amorphous carbon surface.[14] In con-
trast with the fresh sample, there was a gradual in-
crease in propane conversion for all of the annealed
catalysts, which suggests that the active sites were
[a]
[d]
Sample
SBET
Vtotal
sp2 Fraction ID/IG
[%]
Elemental analysis [wt%]
[m2 gÀ1
]
[cm3 gÀ1
]
O
H
ND
313
345
386
450
1.27
1.46
1.55
1.81
27[b]/24[c]
51[b]/60[c]
71[b]/80[c]
94[b]/94[c]
0
0.7
9.8
3.4
0.4
0.1
0.78
0.72
0.06
0.03
ND-1000
ND-1300
ND-1500
0.54
0.36
[a] Total pore volumes, measured at a relative pressure of 0.99. [b] Estimated from
electron energy-loss spectroscopy. [c] Calculated by deconvolution of XPS C1s spectra
(Figure S1). [d] Ratio of D band to G band intensities obtained from Raman spectra.
disordered carbon layer (Figure 1a). IR spectra revealed that
carboxyl and anhydride were the predominant functional
groups on the surface. ND-1000 displayed a highly defective
and curved graphitic surface with one or two graphene layers
(Figure 1b). A certain amount of oxygen detected by elemental
analysis resulted from surface reoxidation by exposure to air
after annealing. For ND-1300, a well-defined graphitic multilay-
er that enclosed the diamond core could be found (Figure 1c),
and the lower defect degree resulted in a significantly de-
creased oxygen content. Upon further annealing at 15008C
(Figure 1d), all the nanoparticles were transformed into spheri-
cal carbon onions with a highly ordered sp2-hybridized graphit-
ic structure and an almost oxygen-free surface. Notably, ND-
1500 still contained a small portion of sp3-hybridized carbon
atoms (Table 1), which may originate from the defects present
on the onion shell.[12] N2 physisorption results showed that the
generated in situ on the surface defects of the graphitic
carbon under the reaction conditions. For different annealed
catalysts, the activation process was prolonged with the in-
creasing annealing temperature, which indicates that the rate
of the generation of active sites was inversely proportional to
the degree of defects of the graphitic shell. Similarly, there was
also a gradual increase in propene selectivity in the initial
stage of the catalytic test (Figure 2b). A remarkably increased
selectivity to propene from 14 to 50% was found on ND-1500,
which implied that the active species with a highly selective
alkane activation may be formed on the graphitic onion-like
carbon structure during reaction.
Cyclic catalytic reaction and temperature-programmed de-
sorption (TPD) experiments were performed to obtain more
detailed information on the removal and regeneration of
active oxygen species during reaction (Figure 3a). It can be ob-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2270 – 2275 2271

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
Figure 3. a) ODH/TPD cycles of the catalytic oxidation of propane catalyzed
by ND-1000. b) In situ DRIFT spectra of ND-1000 in 2%C3H8/O2/He at 4508C.
The difference spectra were obtained by subtraction of the ND background
spectrum in helium.
Figure 2. Catalytic performance of ND-derived catalysts as a function of time
on stream, a) propane conversion and b) propene selectivity.
served that after the consecutive TPD to remove all the surface
oxygen groups, an activation process with relatively low initial
conversions in the second and third reactions appeared, which
suggests that the surfaces of the annealed catalysts are highly
active and the active oxygen species can be regenerated
quickly and repeatedly under the reaction conditions. Addi-
tionally, it can be found that with (the first reaction) or without
initial oxygen groups (the second and third reactions) the cata-
lyst showed a negligible difference in selectivity and reactivity
at the steady state, which indicates that the activity of the
ODH reaction is dependent on these in situ generated active
sites and less related to the initial presence of oxygen groups.
TPD results showed a similar COx desorption peak after each
reaction (Figure S2), which confirms the identical chemical
nature of the regenerated oxygen species during cyclic catalyt-
ic tests.
of the absorbed propane. CO2 that resulted from the oxidative
reaction was detected at around n˜ =2300 cmÀ1. Active oxygen
groups can be generated promptly in situ when the catalyst
surface is exposed to the oxidative atmosphere, as confirmed
by the appearance of C=O vibration bands at n˜ =1760 cmÀ1
with gradually increased intensity. In addition, there was a new
band at n˜ =882 cmÀ1, which was not attributed to either reac-
tants or products in the dehydrogenation process. Indeed, it is
a characteristic band of the out-of-plane bending vibration
mode of propanol,[16] which indicates the formation of hydrox-
yl-like intermediates from the reduction of active C=O groups.
However, the shift of the ketonic C=O band involved in the
redox process was not observed clearly and it may overlap
with the broad band of the carbonyl groups. For the fresh ND
catalyst, the same reaction pathway was also observed in
in situ DRIFT spectroscopy (Figure S3). The difference compared
with the annealed catalysts was that, as a result of the pres-
ence of an unstable amorphous carbon surface, oxygen
groups were found to desorb or decompose during the reac-
tion, as demonstrated by the negative vibration bands of C=O
centered at n˜ =1840 cmÀ1 and CÀO in the range of n˜ =1000–
1300 cmÀ1. This observation is consistent with the gradually
decreased initial catalytic activity on fresh ND.
Direct evidence for the evolution of active oxygen species
can be obtained from the in situ time-resolved diffuse-reflec-
tance infrared Fourier transform (DRIFT) spectra, which were
measured under the reaction conditions (Figure 3b). The differ-
ence spectra obtained by subtracting the ND background
spectrum were used to reflect the development of the oxygen
groups during reaction.[15] The bands in the n˜ =2800–
3100 cmÀ1 region correspond to the CÀH stretching vibration
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2270 – 2275 2272

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
sorption peaks that correspond to the ketone-like groups shift-
ed to a higher temperature with the increasing annealing tem-
perature (Figure S4c–d), which indicates that ketonic carbonyl
groups with a high stability were generated on the recon-
structed graphitic surface. To obtain a more quantitative analy-
sis on the evolution of each oxygen group on different nano-
structures, CO and CO2 desorption profiles of the used cata-
lysts were further deconvoluted based on the methods report-
ed in many studies,[17a,18] and the results are shown in Fig-
ure S5. Lactone groups that released CO2 at a relatively low
temperature were formed predominantly after the reaction,
whereas the increase of the amount of CO desorption was at-
tributed mainly to the formation of phenol/ether and ketone
groups, in accordance with the DRIFT spectra.
The surface ketonic carbonyl groups generally play a key
role in the ODH reaction.[6] The relationship between the
amount of ketone groups, the sp2-hybridized carbon fraction
of catalysts, and the catalytic performances is given in Fig-
ure 5a. The catalytic activity was inconsistent with the de-
crease in the amount of ketonic carbonyls and the increase of
the sp2-bonded carbon fraction. Fresh ND exhibited a lower ac-
tivity than ND-1000, although it had the maximum amount of
ketones. This means that the catalytic activity of ODH was not
simply determined by the content of active ketone groups.
Indeed, with the increasing annealing temperature, the TPD
curves after catalysis showed the increased desorption temper-
ature of ketones, which indicates a structural effect on the for-
mation of active sites. The rate of propane consumption was
further normalized by the amount of ketone groups to reflect
the reactive capacity of the active sites on different hybridized
nanostructures. With the increase of the annealing tempera-
ture, the consumption rate catalyzed by a single ketone in-
creased gradually (Figure 5b). This suggests that the in situ
generated ketone groups on a well-ordered graphitic onion-
like carbon structure were more active than those on a poorly
organized graphitic surface. This structural effect was also con-
firmed by Liu et al.[2c] in the ODH of n-butane, and they found
that the sp3-to-sp2 lattice rearrangement of ND surface was fa-
vorable for the selective generation of active ketonic carbonyl
groups, which could enable high selectivity to butenes. Inter-
estingly, there is a corresponding linear relationship between
the sp2-hybridized carbon fraction and the reactivity of oxygen
species (Figure 5b). Thus it is reasonable to assume that the re-
active properties of active oxygen species can be regulated by
the sp2-hybridized carbon fraction of the catalysts. The nature
of the oxygen groups should be related to the different local
environments as a consequence of structural transformation. A
plausible explanation for this effect is that the increase of the
sp2-hybridized carbon fraction of the inner layers may provide
a greater contribution to the average p-electron density of the
outer shells, which can improve the nucleophilic properties of
the anchored oxygen functional groups and lead to a high
selectivity.[19]
Figure 4. a) DRIFT spectra of used catalysts. b) Calculated oxygen content of
used catalysts based on the desorbed area of CO and CO2 during TPD, the
inset gives the CO/CO2 ratios before and after ODH reaction.
DRIFT spectra of the used catalysts (Figure 4a) showed the
identical chemical nature of the carbonyl groups at n˜ =
1756 cmÀ1 but with different intensity. These groups originated
mainly from the chemisorption of O2 during the reaction and
remained thermally stable on the catalyst surface under oxida-
tive conditions. The variation in intensity suggested that the
sp2@sp3 core–shell nanostructures had a great impact on the
formation of oxygenated functionalities. A better quantitative
analysis of the oxygen functional groups for different hybrid-
ized nanostructures can be obtained from the TPD curves (Fig-
ure S4). In general, CO desorption is attributed mainly to
phenol, ethers, and carbonyl/quinone species, whereas CO2 de-
sorption results from carboxylic anhydride at low temperatures
and lactones at higher temperatures.[17] A remarkable increase
in the oxygen content for all catalysts after ODH reaction
caused by the in situ generation of oxygen species is shown in
Figure 4b. Although the total amount of oxygen groups
formed decreased with the increasing annealing temperature
because of the lower concentration of defects, there was a sig-
nificantly increased ratio of CO/CO2, which suggests that the
nonacidic groups that desorb CO, such as ketone groups, were
formed preferentially on the defects of catalysts annealed at
a higher temperature in the ODH reaction. Also, the CO de-
The conversion–selectivity relationship achieved by different
nanocarbons can give us direct evidence for the capability of
the catalyst in the ODH reaction, in which the annealed cata-
lysts were initially oxidized at 4508C to saturate all defects by
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2270 – 2275 2273

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
which indicates that the unique curved and strained graphitic
surface of ND is indispensable for selective alkane activation.
Conclusions
Nanodiamond (ND) samples annealed in the range of 1000–
15008C were used as model catalysts to study the effect of
structure on the evolution and chemical nature of the oxygen
functional groups in the oxidative dehydrogenation reaction
(ODH). The activation process with different levels over all cata-
lysts was observed in the ODH of propane. We studied in situ
reactions and characterization data and found that the initial
oxygen groups present on the ND surface had little influence
on the catalytic performances. The unstable oxygen groups
tended to desorb or decompose under the reaction conditions,
and active oxygen species can be generated in situ by the
chemisorption of O2 and are involved directly in the catalytic
dehydrogenation process that follows a redox mechanism.
For different hybridized nanostructures, the same type of
generated active groups showed a different catalytic capability,
which can be regulated by the sp2-hybridized carbon fraction
of ND. The sp2-hybridized onion-like carbon structure favors
the generation of highly reactive oxygen species strongly,
which improves the selectivity for the ODH reaction efficiently.
Therefore, this work may lead to a further understanding of
the ODH reaction process catalyzed by carbon materials and
provide possibilities for the controlled modification of
nanocarbons.
Experimental Section
Catalyst preparation
The ND used in this study (high purity grade) was supplied by Bei-
jing Grish Hitech Co. China and it was purified from the black
powder produced by explosive detonation using a nitric acid/sulfu-
ric acid/fuming sulfuric acid mixture. The diamond powder thus
obtained has a light brown color with particle sizes of ꢀ3–10 nm
and a phase purity of powder >98%. The annealing treatments
were conducted as follows: the fresh ND was heated to 1000,
1300, or 15008C at a rate of 58CminÀ1, kept for 4 h under a He
flow, and then cooled to RT prior to exposure to air. These samples
were defined as ND-1000, ND-1300, and ND-1500, respectively.
Figure 5. a) Catalytic performance versus sp2 carbon calculated by electron
energy-loss spectroscopy and content of ketonic carbonyl groups. The con-
tent of ketones was based on the TPD quantitative analysis. b) Propane con-
sumption rate normalized based on the specific content of ketones versus
the sp2 carbon fraction. c) The C3H6 selectivity versus C3H8 conversion curves
obtained at different temperatures for ND-derived catalysts and CNTs.
Characterization
BET and microporous surface area analysis were determined by N2
physisorption at À1968C by using a Micrometrics ASAP 2020 in-
strument. The samples were outgassed at 1508C for 12 h prior to
the isotherm measurement. HRTEM was performed by using a FEI
Cs-corrected Titan 80–300 microscope and a FEI Tecnai G2 F20 mi-
croscope. X-ray photoelectron spectroscopy (XPS) measurements
were performed by using an ESCALAB 250 instrument with AlKa ra-
diation (1489.6 eV).
forming oxygen groups (Figure 5c). It is remarkable that, at
similar conversions, ND-1500 displayed the highest selectivity
to propene, almost 20% higher than the initial ND catalyst.
This evidences that the ketonic carbonyl groups generated on
a well-graphitized surface are highly active and can improve
the selectivity for the ODH reaction efficiently. In addition,
compared with traditional sp2-hybridized carbon nanotubes
(CNTs), approximately 30% higher selectivity can be found,
IR studies were conducted by using a Thermo Nicolet iZ10 FTIR
system using a DRIFT cell that had been modified extensively to
allow in situ treatments up to 5008C under flowing gases. The
spectra were recorded in the 650–4000 cmÀ1 range with 128 scans
at a resolution of 4 cmÀ1
.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2270 – 2275 2274

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[1] 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) H. Yu, F. Peng, J.
Tan, X. W. Hu, H. J. Wang, J. A. Yang, W. X. Zheng, Angew. Chem. Int. Ed.
2011, 50, 3978–3982; Angew. Chem. 2011, 123, 4064–4068.
[2] a) J. Zhang, X. Liu, R. Blume, A. Zhang, R. Schlogl, D. S. Su, Science 2008,
322, 73–77; b) J. Zhang, D. S. Su, R. Blume, R. Schlogl, R. Wang, X. Yang,
A. Gajovic, Angew. Chem. Int. Ed. 2010, 49, 8640–8644; Angew. Chem.
2010, 122, 8822–8826; c) X. Liu, B. Frank, W. Zhang, T. P. Cotter, R.
Schlogl, D. S. Su, Angew. Chem. Int. Ed. 2011, 50, 3318–3322; Angew.
Chem. 2011, 123, 3376–3380.
The surface-oxygenated functionalities were characterized by TPD.
The conditions of the TPD tests were consistent with those of the
catalytic reaction. The sample (150 mg) was first heated to 4508C
and kept for 15 min in a He flow (50 mLminÀ1), then heated to
9508C with a ramp rate of 108CminÀ1 and held for 1 h to remove
the functional groups completely. The signals of released CO and
CO2 were measured on-line by an Omnistar mass spectrometer.
The integration of the peak areas of evolved gases was quantified
based on a calibration curve built by using calcium oxalate
(CaC2O4·H2O) as a standard sample.
[3] K. P. Gong, F. Du, Z. H. Xia, M. Durstock, L. M. Dai, Science 2009, 323,
760–764.
[4] T. J. Bandosz, Carbon Materials for Catalysis (Eds.: P. Serp, J. L. Figueire-
do), Wiley-VCH, Weinheim, 2009, pp. 45–56.
Catalytic tests
[5] J. L. Figueiredo, M. F. R. Pereira, Catal. Today 2010, 150, 2–7.
[6] a) M. F. R. Pereira, J. J. M. Orfao, J. L. Figueiredo, Appl. Catal. A 1999, 184,
153–160; b) J. A. Maciꢁ-Agullꢂ, D. Cazorla-Amoros, A. Linares-Solano, U.
Wild, D. S. Su, R. Schlçgl, Catal. Today 2005, 102, 248–253.
[7] a) D. Su, N. I. Maksimova, G. Mestl, V. L. Kuznetsov, V. Keller, R. Schlçgl,
N. Keller, Carbon 2007, 45, 2145–2151; b) J. J. Delgado, X. Chen, J. P.
Tessonnier, M. E. Schuster, E. Del Rio, R. Schlogl, D. S. Su, Catal. Today
2010, 150, 49–54.
[8] N. R. Greiner, D. S. Phillips, J. D. Johnson, F. Volk, Nature 1988, 333, 440–
442.
[9] A. E. Aleksenskii, M. V. Baidakova, A. Y. Vul’, V. I. Siklitskii, Phys. Solid State
1999, 41, 668–671.
The ODH was performed in a quartz fixed-bed reactor under at-
mospheric pressure. The sample (250 mg) was preheated to 4508C
in He, and 2% C3H8/1% O2/He (15 mLminÀ1) was fed to the reactor.
The reaction products were analyzed by using an Agilent 7890 A
gas chromatograph equipped with a flame ionization detector
(FID) for hydrocarbons and a thermal conductivity detector (TCD)
for inorganic components. Blank experiments showed that the re-
action rates were negligible without the carbon catalyst. In all
tests, carbon mass balances were within 100Æ0.5%.
Cyclic TPD and catalytic tests were performed using ND-1000 in
the reactor, and the reaction products were also analyzed by GC. A
mass spectrometer was used for the TPD characterization of sur-
face functionalities. The whole experiment was performed follow-
ing these steps: firstly, the ODH reaction was performed for 16 h.
After reaction, the reactant input was replaced with helium
(30 mLminÀ1), and the temperature was rapidly decreased to RT.
TPD was performed with a heating ramp of 108CminÀ1, and kept
at the maximum temperature of 9508C for 1 h. After that, the tem-
perature was reduced to 4508C, and the ODH reaction was per-
formed again by switching to the reaction feed. All of the above
processes were repeated three times, and the catalyst was not re-
moved from the reactor or exposed to air.
[10] a) Y. V. Butenko, V. L. Kuznetsov, E. A. Paukshtis, A. I. Stadnichenko, I. N.
Mazov, S. I. Moseenkov, A. I. Boronin, S. V. Kosheev, Fullerenes Nanotubes
Carbon Nanostruct. 2006, 14, 557–564; b) F. Y. Xie, W. G. Xie, L. Gong,
W. H. Zhang, S. H. Chen, Q. Z. Zhang, J. Chen, Surf. Interface Anal. 2010,
42, 1514–1518; c) O. O. Mykhaylyk, Y. M. Solonin, D. N. Batchelder, R.
Brydson, J. Appl. Phys. 2005, 97, 074302; d) Z. J. Qiao, J. J. Li, N. Q. Zhao,
C. S. Shi, P. Nash, Scr. Mater. 2006, 54, 225–229.
[11] R. Wang, X. Y. Sun, B. S. Zhang, X. Y. Sun, D. S. Su, Chem. Eur. J. 2014,
DOI: 10.1002/chem.201400018.
[12] S. Tomita, M. Fujii, S. Hayashi, K. Yamamoto, Chem. Phys. Lett. 1999, 305,
225–229.
[13] V. L. Kuznetsov, Y. V. Butenko, Nato. Sci. Ser. Ii-Math. 2005, 192, 199–216.
[14] A. Rinaldi, J. Zhang, B. Frank, D. S. Su, S. B. Abd Hamid, R. Schlçgl, Chem-
SusChem 2010, 3, 254–260.
[15] A. Pantazidis, C. Mirodatos, Stud. Surf. Sci. Catal. 1996, 101, 1029–1038.
[16] E. K. Plyler, J. Res. Natl. Bur. Stand. 1952, 48, 281–286.
[17] a) J. L. Figueiredo, M. F. R. Pereira, M. M. A. Freitas, J. J. M. Orfao, Carbon
1999, 37, 1379–1389; b) H. P. Boehm, Carbon 2002, 40, 145–149.
[18] a) J. H. Zhou, Z. J. Sui, J. Zhu, P. Li, C. De, Y. C. Dai, W. K. Yuan, Carbon
2007, 45, 785–796; b) N. Li, X. L. Ma, Q. F. Zha, K. Kim, Y. S. Chen, C. S.
Song, Carbon 2011, 49, 5002–5013; c) J. L. Figueiredo, M. F. R. Pereira,
M. M. A. Freitas, J. J. M. Orfao, Ind. Eng. Chem. Res. 2007, 46, 4110–4115.
[19] a) C. Portet, G. Yushin, Y. Gogotsi, Carbon 2007, 45, 2511–2518; b) V. L.
Kuznetsov, Yu. V. Butenko, A. L. Chuvilin, A. I. Romanenko, A. V. Okotrub,
Chem. Phys. Lett. 2001, 336, 397–404.
Acknowledgements
The authors sincerely acknowledge discussions and assistance by
Y. Lin and J. Wang. This work was supported by the Foundation
from MOST (2011CBA00504), NSFC of China (21133010, 21103203,
51221264, 21261160487, 21203215), and the “Strategic Priority
Research Program” of the Chinese Academy of Sciences, Grant
No. XDA09030103.
Keywords: alkanes
·
carbon
·
dehydrogenation
·
Received: March 3, 2014
nanostructures · oxygen
Published online on July 10, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2270 – 2275 2275