Isobutane dehydrogenation over the mesoporous Cr2O 3/Al2O3 catalysts synthesized from a metal-organic framework MIL-101

Article abstract of DOI:10.1016/j.apcata.2013.02.018

The reactivity of isobutane dehydrogenation over a series of non-ordered mesoporous chromia/alumina (Cr₂O₃/Al₂O ₃) catalysts with large specific surface area (149.4-381.6 m ² g^-1) and high pore volume (0.77-1.24 cm³ g^-1), synthesized using a metal-organic framework MIL-101 as a molecular host and chromium precursor, aluminium isopropoxide (Al(i-OC ₃H₇)₃) as the aluminium precursor, were studied in detail. The chromium species were highly dispersed over the catalyst with chromia loading up to 10 wt.%. The specific surface area of the catalyst decreased, whereas the amount of surface Cr³⁺ species and the mole ratio of Cr³⁺ and Cr⁶⁺ species (Cr³⁺/Cr ⁶⁺) increased with the increasing chromia loadings (5-25 wt.%) and calcination temperature (500-900 °C), respectively. The addition of potassium to the catalyst system greatly promoted isobutene selectivity and catalyst stability. The catalyst with 1.5 wt.% K₂O and 10 wt.% Cr₂O₃ loadings calcined at 800 °C was found to exhibit the highest isobutane conversion 60.1% with the isobutene selectivity up to 93.2% among all the catalysts. The maintainable catalytic reactivity demonstrated the high stability of the catalyst in ten dehydrogenation- regeneration cycles. Moreover, it was proposed that the Cr³⁺ species was mainly the active site and catalytic selectivity was depended on the surface Cr³⁺/Cr⁶⁺ value over the catalyst. The catalyst presented much more stable dehydrogenation activity compared with the conventional catalyst. Consequently, this study presents a feasible way to facile synthesis of the mesoporous MOF-derived Cr₂O₃/Al₂O ₃ catalysts with high stability and good catalytic reactivity over isobutane dehydrogenation.

Full text of DOI:10.1016/j.apcata.2013.02.018

Applied Catalysis A: General 456 (2013) 188–196
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Isobutane dehydrogenation over the mesoporous Cr O /Al O
3
2
3
2
catalysts synthesized from a metal-organic framework MIL-101
Huahua Zhaoa,b, Huanling Songa,c, Leilei Xua,b, Lingjun Choua,c,∗
a
State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, PR
China
b
University of Chinese Academy of Sciences, Beijing 100049, PR China
Suzhou Institute of Nano-Tech and Nano-Bionics, Suzhou 215123, PR China
c
a r t i c l e i n f o
a b s t r a c t
Article history:
The reactivity of isobutane dehydrogenation over a series of non-ordered mesoporous chromia/alumina
Received 23 November 2012
Received in revised form 1 February 2013
Accepted 16 February 2013
Available online xxx
2
−1
(
(
Cr2O3/Al2O3) catalysts with large specific surface area (149.4–381.6 m g ) and high pore volume
0.77–1.24 cm g ), synthesized using a metal-organic framework MIL-101 as a molecular host and
3
−1
chromium precursor, aluminium isopropoxide (Al(i-OC3H7)3) as the aluminium precursor, were stud-
ied in detail. The chromium species were highly dispersed over the catalyst with chromia loading up to
3
+
1
0 wt.%. The specific surface area of the catalyst decreased, whereas the amount of surface Cr species
Keywords:
3
+
6+
3+
6+
and the mole ratio of Cr and Cr species (Cr /Cr ) increased with the increasing chromia loadings
5–25 wt.%) and calcination temperature (500–900 C), respectively. The addition of potassium to the cat-
Chromia/alumina
Isobutane dehydrogenation
Mesoporous
Metal-organic framework
MIL-101
◦
(
alyst system greatly promoted isobutene selectivity and catalyst stability. The catalyst with 1.5 wt.% K2O
◦
and 10 wt.% Cr2O3 loadings calcined at 800 C was found to exhibit the highest isobutane conversion 60.1%
with the isobutene selectivity up to 93.2% among all the catalysts. The maintainable catalytic reactivity
demonstrated the high stability of the catalyst in ten dehydrogenation-regeneration cycles. Moreover,
3
+
it was proposed that the Cr species was mainly the active site and catalytic selectivity was depended
3
+
6+
on the surface Cr /Cr value over the catalyst. The catalyst presented much more stable dehydrogena-
tion activity compared with the conventional catalyst. Consequently, this study presents a feasible way
to facile synthesis of the mesoporous MOF-derived Cr2O3/Al2O3 catalysts with high stability and good
catalytic reactivity over isobutane dehydrogenation.
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
activity and selectivity and suppressing the coke formation during
the catalytic dehydrogenation reaction.
Catalytic dehydrogenation of isobutane over the chro-
Supported Cr O /Al O catalyst has been applied for the isobu-
2
3
2
3
mia/alumina (Cr O /Al O ) catalyst is an industrially important
tane dehydrogenation on a commercial scale for years [10].
Currently, this catalyst system occupies more than half of the world
market of the commercial catalysts for paraffin dehydrogenation
[11]. It is well known that the nature of surface chromia is related
to the catalytic activity and selectivity. The speciation of surface
chromia is strongly depended on the type of the support, the
chromium content and calcination treatment [1,2,12–15]. De Rossi
et al. reported that Cr6+ species with minor amount of Cr species
2
3
2
3
route for producing isobutene, a vital component for the synthesis
of MTBE (methyl tert-butyl ether) and ETBE (ethyl tert-butyl ether),
which are increasingly demanded as octane number boosters for
unleaded gasoline in the worldwide recently [1–6]. Dehydrogena-
tion is typically performed at relatively high temperatures and
low pressures due to thermodynamic requirements. However,
reaction conditions, such as high temperature, usually facilitate
the thermal cracking and coke formation, thus result in reduced
isobutane conversion, decreased isobutene selectivity and severe
deactivation of the catalyst [7–9]. It is therefore the key issue to
develop catalyst with the ability of possessing high and stable
5+
6+
5+
were produced on ␥-Al O and silica carriers, whereas Cr and Cr
2
3
species in comparable amounts on the zirconia support [13]. Cavani
3+
et al. observed that the relative concentration of surface Cr and
6+
Cr species increased with the increasing chromia loading. They
3+
also found that only Cr species were produced over the surface of
the reduced catalyst [2]. Besides, the researchers have been devoted
to improving the catalytic dehydrogenation activity and selectivity
as well as the catalyst stability. It has been reported that the addi-
tion of alkali metal, such as potassium, to the oxide catalyst system
usually promotes the catalytic selectivity of olefins and improves
∗ Corresponding author at: State Key Laboratory for Oxo Synthesis and Selec-
tive Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences,
Lanzhou 730000, PR China. Tel.: +86 0931 4968066; fax: +86 0931 4968129.
E-mail address: ljchou@licp.cas.cn (L. Chou).
0
926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.02.018

H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
189
the resistance of the catalyst to deactivation [2,6,16–22]. Many
preparation methods for the Cr O /Al O catalysts described in the
temperature) with different Cr O3 loadings (5–25 wt.%) were pre-
pared through controlling the addition amount of MIL-101.
2
2
3
2
3
literatures, such as incipient wetness impregnation, atomic layer
epitaxy (ALE), co-precipitation method and evaporation-induced
self-assembly (EISA) strategy, have been explored over the alkanes
dehydrogenation [2,7,15,23,24]. However, the utilization of the
method that the precursor containing aluminium source impreg-
nated into the Cr-based framework to prepare the Cr O /Al O
catalyst for isobutane dehydrogenation has been paid little atten-
tion so far. The similar reports for this method were employed to
prepare the nickel catalysts in the reaction of methane decomposi-
tion [25–27]. This method may be contributable to a high dispersion
of chromium species over the support.
To examine the additive K O over the reactivity of isobutane
2
dehydrogenation, KOH (0.04 g) together with Al(i-OC H7) was dis-
3
3
solved in EtOH solution according to the preparation steps of the
catalyst 10Cr O /Al O (600) mentioned above to obtain the cata-
2
3
2
3
lyst 1.5 K O-10Cr O /Al O (600) with 1.5 wt.% K O loading.
2
2
3
2
3
2
In order to investigate the calcination temperature over the cat-
alytic activity and selectivity of isobutane dehydrogenation, the
catalyst 1.5 K O-10Cr O /Al O (T) (T stands for the calcination
2
3
2
3
2
2
3
2
3
◦
◦
◦
◦
temperatures) was calcined at 500 C, 600 C, 700 C, 800 C and
900 C, respectively.
◦
The reference alumina supported chromia catalyst 1.5 K O-
2
MIL-101 (Cr F(H O) O(BDC) ·nH O, n ≈ 25, MIL = Matérial
10Cr O /Al O (800)-R was prepared by impregnation method and
3
2
2
3
2
2
3
2
3
Institut Lavoisier, BDC = 1,4-benzenedicarboxylate) is one of the
porous Cr-based metal-organic frameworks (MOFs). It has a
large Brunauer–Emmett–Teller (BET) specific surface area about
described as follows: KOH (0.04 g) and Cr(NO ) ·9H O (0.90 g) were
3
3
2
dissolved in EtOH solvent (100 mL). After adding ␥-Al O (1.53 g),
2
3
◦
the mixture was stirred at 50 C to evaporate slowly the solvent to
form a green paste. The following preparation steps were the same
as those of the catalyst 1.5 K O-10Cr O /Al O (800).
2
−1 −1
3
4
100 m g and high pore volume about 2.0 cm g [28–31]. Its
two large mesopores with 2.9 and 3.4 nm pore diameters together
with its considerable chromium content (21.7%) stimulate us
2
2
3
2
3
to prepare the mesoporous Cr O /Al O catalyst using it as a
2.3. Catalytic tests
2
3
2
3
molecular flask and chromium precursor [32–35]. In the current
work, the mesoporous Cr O /Al O catalyst with large specific
The reactivity of the each catalyst over isobutane dehydrogena-
tion was carried out in a continuous flow reactor consisting of a
fixed-bed quartz tube at 600 C. The each catalyst (0.60 g, 60–80
mesh) was loaded in the reactor supported on the quartz sand,
then heated to the set temperature under the N2 flow. The isobu-
tane was fed into the reactor during 5 min. The reaction products
were analyzed for 2 h time-on-stream (TOS) with gas chromatogra-
phy. The original green colour over the fresh catalysts was changed
to the black one after the isobutane dehydrogenation. In case of
2
3
2
3
surface area and high pore volume was synthesized by utilizing the
MIL-101 framework as a molecular host and chromium precursor,
aluminium isopropoxide (Al(i-OC H7) ) as the aluminium precur-
◦
3
3
sor. The chromia loadings, addition of potassium oxide (K O) and
2
calcination temperature of the catalyst were investigated in detail
over the isobutane dehydrogenation. The high isobutene selectiv-
ity 93.2% at the high isobutane conversion of 60.1% was obtained
in the experiment. In order to study the stability and regenerative
ability of the catalyst, the ten dehydrogenation-regeneration
cycles were carried out.
the ten dehydrogenation-regeneration cycles, the catalyst 1.5 K O-
2
◦
10Cr O /Al O (800) was regenerated under air at 650 C for 2 h and
2
3
2
3
cooled down to room temperature, then utilized to the next cycle.
2
. Experimental
2.4. Characterizations
2
.1. Materials
Powder X-ray diffraction measurements (PXRD) of the products
were performed on an X-ray diffractometer (PANalytical) by using
◦
1
,4-Benzene dicarboxylic acid (H BDC, C H O , ≥98.0%) and
Ni-filtered Cu K␣ radiation (ꢀ = 0.15418 nm) from 0.6 to 5.0 (small
2
8
6
4
◦
aluminium isopropoxide (Al(i-OC H7) ,
chased from Sigma–Aldrich (USA). Chromium nitrate nonahydrate
≥
98.0%) were pur-
angle range) and 20.0 to 80.0 (wide angle range).
3
3
The N2 adsorption–desorption isotherms were detected on a
◦
(
Cr(NO ) ·9H O, ≥99.0%), hydrofluoric acid (HF, ≥40.0%), potas-
Quantachrome autosorb iQ gas-sorption apparatus at −196 C. All
3
3
2
◦
sium hydroxide(KOH, ≥82.0%)and absoluteethanol(EtOH, ≥99.7%)
samples except MIL-101 (evacuated at 150 C for 6 h) were acti-
vated by degassing in situ under vacuum at 200 C for 4 h prior to
◦
were from Sinopharm Chemical Reagent Co., Ltd (China). The indus-
2
−1
trial ␥-Al O (SBET = 332.3 m g ) was purchased from Guizhou
Aluminium Corporation (China). All the materials were used as
received from vendors without further purification.
the measurements. BET specific surface areas of the samples were
estimated over the relative pressure range of 0.05–0.30. The pore
size of MIL-101 was calculated according to the Non-Local Density
Functional Theory (NLDFT) and the pore sizes of the catalysts were
calculated based on the Barrett–Joiner–Halenda (BJH) method.
The products were subjected to the transmission election
microscopy (TEM) analysis on the FEI TECNAI G2 instrument oper-
ated at 200 kV.
2
3
2
.2. Catalysts synthesis
MIL-101 was synthesized by the solvothermal method accord-
ing to the previous literature with the exception that the mole ratio
Ultraviolet–visible spectroscopy (UV–vis) of the samples was
collected on a PE Lambda 650S instrument coupled with a diffuse
reflectance accessory.
X-ray photoelectron spectroscopy (XPS) of the products was
carried out on a VG ESCALAB 250 apparatus equipped with Al K␣
and Mg K␣ X-ray source and a hemispheric analyzer. The binding
energies were referenced to the C 1s peak from the carbon surface
deposit at 284.8 eV.
of n(Cr(NO ) ·9H O:n(H BDC):n(HF): n(H O) was 1:1:0.5:280 [36].
3
3
2
2
2
The as-synthesized MIL-101 was degassed under reduced pressure
◦
at 150 C for 12 h to remove the guest molecules. The pretreated
MIL-101 (0.56 g) was added into absolute ethanol (EtOH) solution
−
1
containing Al(i-OC H7)3 (0.30 mol L , 100 mL) and impregnated
3
◦
by ultrasonic for 1.5 h. Afterwards, the mixture was stirred at 50 C
to evaporate slowly the solvent to form a green paste. The paste
◦
was dried at 120 C overnight to obtain a bright green powder,
Temperature programmed reduction (TPR) measurements for
the products with H2 as the reducing agent were monitored on
a thermal conductivity detector (TCD) cell (TP-5080). Prior to the
◦
which was calcined at 600 C for 4 h in air to obtain the catalyst,
denoted as 10Cr O /Al O (600). The catalysts xCr O /Al O (600)
2
3
2
3
2
3
2
3
◦
(
x wt.% stands for the weight ratio of Cr O3 loading in the cata-
reduction steps, samples were pretreated at 300 C for 0.5 h in flow-
2
−
1
lysts and the number in parentheses represents for the calcination
ing He (30 mL min ) to remove any moisture and other impurities

1
90
H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
Table 1
2
−1
the catalyst 10Cr O /Al O (600) slightly declined to 334.8 m g
2
3
2
3
The textural data of the catalysts.
3
−1
and 1.04 cm g over the one 1.5 K O-10Cr O /Al O (600). Nev-
ertheless, the addition of K O had no remarkable influence over the
2
2
3
2
3
SBET (m2
g
−1
)
Vp (cm3
g
−1
)
Dp (nm)
Sample
2
pore size (9.54 nm) of the catalyst, which indicated that introduc-
tion of K2O caused no pore plugging of the mesoporous Cr2O3/Al2O3
catalyst.
5
1
1
2
2
1
1
1
1
1
1
1
1
1
Cr2O3/Al2O3(600)
0Cr2O3/Al2O3(600)
5Cr2O3/Al2O3(600)
0Cr2O3/Al2O3(600)
381.6
375.6
324.6
267.1
237.0
337.2
334.8
264.7
197.6
149.4
190.9
177.9
172.1
234.3
1.21
1.24
1.21
0.78
1.09
1.17
1.04
1.08
0.87
0.77
0.65
0.55
0.53
0.29
9.52
9.55
12.32
7.79
12.30
9.54
9.54
12.29
16.93
17.22
12.28
12.29
9.55
In case of the series catalysts 1.5 K O-10Cr O /Al O (T) calcined
2
2
3
2
3
5Cr2O3/Al2O3(600)
at different temperatures, 1.5 K O-10Cr O /Al O (500) possessed
2
2
3
2
3
.5 K2O-10Cr2O3/Al2O3(500)
.5 K2O-10Cr2O3/Al2O3(600)
.5 K2O-10Cr2O3/Al2O3(700)
.5 K2O-10Cr2O3/Al2O3(800)
.5 K2O-10Cr2O3/Al2O3(900)
.5 K2O-10Cr2O3/Al2O3(800)
.5 K2O-10Cr2O3/Al2O3(800)
2
−1
the largest BET specific surface area (337.2 m g ) and highest pore
3
−1
volume (1.17 cm g ). It was observed that both the BET specific
surface area and pore volume of the catalyst became smaller and
the pore size gradually grew larger at the calcination tempera-
ture from 500 C to 900 C. Moreover, the pore size distribution
of the catalyst grew broader with increasing calcination tem-
perature (especially calcined at 800 C and 900 C, respectively).
The phenomenon was probably ascribed to that a small portion
of the pores went collapsed at high calcination temperature. By
a
◦
◦
b
.5 K2O-10Cr2O3/Al2O3(800)c
.5 K2O-10Cr2O3/Al2O3(800)-R
◦
◦
3.41
a,b,c
stand for the corresponding spent, first and tenth regenerated catalyst, respec-
tively.
2
−1
contrast with the reference catalyst (234.3 m g ), the catalyst
1
1
.5 K O-10Cr O /Al O (800) showed a smaller specific surface area
90.9 m g
2 2 3 2 3
that may be present. After cooling the reactor to room temper-
ature, the sample (50 mg) was loaded in a quartz reactor, then
2
−1
.
◦
◦
−1
heated to 800 C at a rate of 10 C min under a flow of 5% H –Ar
2
−
1
(
30 mL min ) gas mixture.
3.2.2. PXRD patterns
The PXRD patterns for the series fresh xCr O /Al O catalysts
2
3
2
3
are presented in Fig. 1A. No diffraction peaks are revealed in the
3
. Results and discussion
◦
small angle range (0.6–5 ) of the series xCr O /Al O catalysts (not
2
3
2
3
show), indicating their non-ordered pore structure. The crystalline
chromia phase was not visible in the PXRD patterns until the chro-
mia loading was up to 15 wt.%, suggesting that the sizes of chromia
particles were very small and the particles were highly dispersed
at least up to 10 wt.% loading on the alumina support [12]. Fur-
3
.1. The preparation of the mesoporous Cr O /Al O catalysts
2
3
2
3
The preparation procedure of the mesoporous Cr O /Al O cat-
2
3
2
3
alysts is suggested in Scheme S1 (See the supporting information).
The Al(i-OC H7) molecules go into the framework of MIL-101 by
adsorption and diffusion with ultrasonic impregnation method to
obtain a composite precursor. Then the precursor is directly cal-
cined at a set temperature by decomposing the organic part to
3
3
◦
◦
thermore, the three broad and weak peaks at 2 ꢁ = 37.5 , 45.6 and
◦
6
6.5 indicated the appearance of the crystalline ␥-Al O phase in
2
3
the support. However, the phase of alumina support was mainly
◦
the amorphous phase calcined at 600 C.
produce the Cr O /Al O catalysts.
2
3
2
3
The PXRD patterns of the fresh catalysts 1.5 K O-
2
1
0Cr O /Al O (T) calcined at different temperatures are described
2
3
2
3
3
.2. The characterizations of the fresh catalysts
in Fig. 1B. Similar to the catalysts above, no peak signals were
observed in the small angle range for the series catalysts (not given),
also implying their non-ordered pore structures. K O phase was
not revealed in the PXRD pattern of the K-doped catalyst, clearly
demonstrating that potassium species were highly dispersed on
the alumina support and had no significant influence over the
3
.2.1. Textural analysis
The N2 adsorption-desorption isotherms and pore size distri-
2
butions for the two series fresh catalysts xCr O /Al O (600) and
1
between adsorption and desorption branches can be observed for
all the catalysts, which demonstrated the existence of mesopores.
Table 1 summarizes the textural data of the catalysts. The aver-
age pore sizes of the catalysts were located between 7.79 nm and
2
3
2
3
.5 K O-10Cr O /Al O (T) are shown in Fig. S1. Hysteresis loops
2 2 3 2 3
phase of the catalyst. Moreover, the crystalline Cr O3 phase was
2
not visible in the PXRD patterns of all the catalysts, indicating
that chromia species were highly dispersed and the calcination
temperature did not affect the dispersion of the chromia species.
Simultaneously, it was observed that the characteristic peak shape
and intensity of ␥-Al O3 phase in the PXRD patterns became
sharper and stronger with the higher calcination temperature,
suggesting that the support calcined at the higher temperature
possessed the pore wall with more crystalline ␥-Al O3 phase.
However, another crystalline phase ␣-Al O3 was visible in the
catalyst 1.5 K O-10Cr O /Al O (900), implying that ␥-Al O phase
was partly transformed to ␣-Al O3 phase at 900 C.
1
7.22 nm, clearly demonstrated its mesoporosity. It can be notice-
ably seen that the xCr O /Al O (600) catalysts with large BET
specific surface areas (237.0–381.6 m g ) and high pore volumes
2
3
2
3
2
−1
2
3
−1
(
0.78–1.24 cm g ) were obtained with this method, which were
much higher than that of the mesoporous Cr O /Al O catalysts
prepared with EISA strategy reported by Shee et al. [24]. For the
series xCr O /Al O (600) catalysts, the BET specific surface area
decreased from 381.6 m g to 237.0 m g as the chromia load-
ing increased from 5 wt.% to 25 wt.%. This was due to that the small
chromia particles gradually aggregated to large ones over the sur-
face of the catalyst with the increase of Cr O loading, thus blocking
2
3
2
3
2
2
2
3
2
3
2
−1
2
−1
2
2
3
2
3
2
3
◦
2
3.2.3. TEM images
2
3
a portion of pores and resulting in the decrease of the specific sur-
face area of the catalyst. The pore size distribution of the catalyst
became broader and the pore size of the catalyst grew larger from
It was very difficult to obtain the clear TEM images due to the
samples were extremely sensitive to the electron beam and the
structure collapsed after a few minutes. However, the TEM images
of the fresh catalysts in Fig. 2a–f can show their pore structures.
In accordance with the N2 adsorption-desorption characteriza-
tion, the pore size appeared to be 9 nm, 10 nm, 12 nm, 17 nm and
17.5 nm over the catalysts 1.5 K O-10Cr O -Al O (T) calcined at
9
.52 nm to 12.32 nm as the Cr O3 loading increased from 5 wt.%
2
to 15 wt.%, which were probably due to that the porous MIL-101
framework had some influence over the pore structure of the cat-
alyst.
2
2
3
2
3
◦
◦
◦
◦
◦
After K O was introduced to the catalyst system, both the BET
500 C, 600 C, 700 C, 800 C and 900 C, respectively. The worm-
like mesopore structure suggested the absence of the long-range
2
2
−1
3
−1
specific surface area 375.6 m g and pore volume 1.24 cm g of

H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
191
◦
◦
◦
◦
◦
Fig. 1. The PXRD patterns of the series fresh catalysts (A) xCr2O3/Al2O3; (B) 1.5 K2O-10Cr2O3/Al2O3(T) calcined at (a) 500 C (b) 600 C (c) 700 C (d) 800 C (e) 900 C (f) the
spent and (g) the tenth regenerated one 1.5 K2O-10Cr2O3/Al2O3(800) (h) the reference one 1.5 K2O-10Cr2O3/Al2O3(800)-R.
◦
◦
◦
◦
◦
Fig. 2. The TEM images of the fresh catalysts (a) 10Cr2O3/Al2O3(600) and 1.5 K2O-10Cr2O3/Al2O3(T) calcined at (b) 500 C (c) 600 C (d) 700 C (e) 800 C (f) 900 C and the
spent ones (g) 10Cr2O3/Al2O3(600) (h) 1.5 K2O-10Cr2O3/Al2O3(800) (i) 1.5 K2O-10Cr2O3/Al2O3(800) after ten dehydrogenation-regeneration cycles.

1
92
H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
uniform pore channels, which were well consistent with their PXRD
results. It was noticeably seen that no extra nanoparticles were
over the catalysts, indicating that potassium and chromia species
were both highly dispersed on all the catalysts and the addition
of K O had no marked influence over the pore structures of the
2
catalysts.
3.2.4. XPS spectra
Table 2 summarizes the XPS data for the studied catalysts. The
Cr 2p3/2 XPS spectra of the samples divided by fitting the dou-
blet peaks are described in Fig. S2. The bands were located at the
binding energy (BE) = 576.5–577.4 eV and 579.4–580.3 eV, which
3
+
6+
were attributed to Cr and Cr species, respectively [2,14,37–40].
As expected, the total amount of surface chromium enhanced
from 1.46% to 4.81% with the increasing chromia loading for the
3
+
6+
series catalysts xCr O /Al O (600). The mole ratio of Cr and Cr
2
3
2
3
3+
6+
(
Cr /Cr ) increased from 1.30 to 2.05 over the catalyst with chro-
mia loading from 5 wt.% to 20 wt.%. Nevertheless, the Cr /Cr
3+
6+
Fig. 3. The UV–vis spectra of the series fresh catalysts 1.5 K2O-10Cr2O3/Al2O3(T).
value decreased to 1.57 over the catalyst 25Cr O /Al O (600). It
2
+
3
2
3
was calculated that the amounts of surface Cr3 and Cr species
6+
both increased with the increasing chromia loading, whereas the
dominant species over the surface of the catalyst with the higher
calcination temperature, as were shown in Table 2. By comparison,
the catalyst 1.5 K2O-10Cr2O3/Al2O3(800) and the reference cata-
amount of Cr species increased more greatly than that of Cr6+
species over the surface of the series catalysts up to 20 wt.% chromia
loading.
3+
3+
6+
lyst 1.5 K2O-10Cr2O3/Al2O3(800)-R presented a similar Cr /Cr
value (1.91 and 1.96, respectively). The surface Cr3+ amount over
the catalyst 1.5 K2O-10Cr2O3/Al2O3(800)-R was 1.56%, which was
more than that (1.29%) of the catalyst 1.5 K2O-10Cr2O3/Al2O3(800).
This meant that the catalyst 1.5 K2O-10Cr2O3/Al2O3(800) had more
The addition of K O to the catalyst system led to the total amount
2
of surface Cr species increasing slightly from 1.78% to 2.01%. The
3+
6+
Cr /Cr value, however, decreased largely from 1.68 to 1.16. It was
3+
observed that the amount of surface Cr species decreased slightly
and the amount of surface Cr6+ species increased dramatically upon
the K O addition to the catalyst, which indicated that K O could
3+
inner type of Cr species than the reference catalyst, indicating that
the catalyst possessed a more stable support structure[11,41]. In a
2
2
6
+
3+
6+
stabilize the Cr species mainly at the expense of the inner ␣-Cr O
word, both Cr and Cr species were observed on the surface of the
fresh catalysts, their different concentrations were mainly depend-
ing on the overall chromia loadings, the promoter K2O addition and
calcination temperature in this experiment.
2
3
phase [2,20].
In case of the series catalysts 1.5 K O-10Cr O /Al O (T), the
2
2
3
2
3
total surface chromium and potassium amounts of the catalyst
◦
gradually increased with the calcination temperature from 500 C
◦
to 900 C, which was probably due to that the abatement in spe-
3.2.5. Diffuse reflectance UV-vis spectra
cific surface area of the catalyst resulted in the decline for the
dispersion of chromium and potassium species over the catalyst.
The diffuse reflectance UV–vis spectra of the fresh catalysts
1.5 K O-10Cr O /Al O (T) calcined at different temperatures are
2
2
3
2
3
3
+
6+
The surface Cr /Cr value was 0.87 over the catalyst 1.5 K O-
plotted in Fig. 3. All spectra exhibited two intense absorption bands
2
6+
2−
6+
1
0Cr O /Al O (500), demonstrating that the amount of Cr
at about 271 nm and 366 nm, contributable to O →Cr charge
2
3
2
3
3+
6+
species was more than that of Cr species over the surface of the
catalyst calcined at 500 C. However, the Cr /Cr value increased
from 1.16 to 2.57 with the calcination temperature increasing
transfer of chromate species, demonstrating the presence of Cr
◦
3+
6+
species on the surface of the catalysts [2,13,39,42,43]. A shoulder
peak at 434 nm was assigned to the d-d transition (A2g→T ) of Cr3+
1g
◦
◦
3+
from 600 C to 900 C, indicating that Cr gradually became the
species in octahedral symmetry, which band may also derive from
Table 2
The XPS data and the results of isobutane dehydrogenation over the catalysts.
Cr3+/Cr6+ (mole ratio)
Activitya (mmol g−1
h
−1
)
TOFb (h−1
)
Catalysts
Surface concn. (mol%)
Cr
Cr3+
5
1
1
2
2
1
1
1
1
1
1
1
1
1
1
Cr2O3/Al2O3(600)
0Cr2O3/Al2O3(600)
5Cr2O3/Al2O3(600)
0Cr2O3/Al2O3(600)
1.46
1.78
2.86
3.54
4.81
1.95
2.01
2.04
1.96
1.96
1.96
3.70
2.04
2.03
2.36
0.83
1.11
1.87
2.38
2.94
0.91
1.08
1.22
1.29
1.41
1.29
3.70
1.41
1.40
1.56
1.30
1.68
1.90
2.05
1.57
0.87
1.16
1.48
1.91
2.57
1.91
–
12.8
18.3
17.2
17.6
17.3
12.9
17.5
21.5
22.4
21.4
29.6
–
1.54
1.65
0.92
0.74
0.59
1.42
1.62
1.76
1.73
1.52
2.30
–
5Cr2O3/Al2O3(600)
.5 K2O-10Cr2O3/Al2O3(500)
.5 K2O-10Cr2O3/Al2O3(600)
.5 K2O-10Cr2O3/Al2O3(700)
.5 K2O-10Cr2O3/Al2O3(800)
.5 K2O-10Cr2O3/Al2O3(900)
.5 K2O-10Cr2O3/Al2O3(800)c
d
.5 K2O-10Cr2O3/Al2O3(800)
e
.5 K2O-10Cr2O3/Al2O3(800)
2.22
2.22
1.96
28.3
29.0
32.2
2.00
2.07
2.06
.5 K2O-10Cr2O3/Al2O3(800)f
.5 K2O-10Cr2O3/Al2O3(800)-Rg
a
b
3
+
c
,
d
,
e
,
f
,
g
A
c
t
i
v
i
t
y
w
a
s
d
e
fi
n
e
d
a
s
m
m
o
l
o
f
i
s
o
b
u
t
a
n
e
c
o
n
v
e
r
t
e
d
p
e
r
g
r
a
m
o
f
c
a
t
a
l
y
s
t
p
e
r
h
o
u
r
.
T
O
F
w
a
s
d
e
fi
n
e
d
a
s
n
u
m
b
e
r
o
f
i
s
o
b
u
t
a
n
e
c
o
n
v
e
r
t
e
d
p
e
r
C
r
s
i
t
e
p
e
r
h
o
u
r
.
The activity and TOF stand for the fresh, spent, the first and tenth regenerated catalyst 1.5 K2O-10Cr2O3/Al2O3(800) and the reference catalyst 1.5 K2O-10Cr2O3/Al2O3(800)-R
−1
over the isobutane dehydrogenation at the GHSV = 1200 h , respectively.

H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
193
Fig. 5. The reactivity of the isobutane dehydrogenation over the series catalysts
xCr2O3/Al2O3.
Fig. 4. The H2-TPR profiles of (a) 10Cr2O3/Al2O3(600) and the series catalysts
◦
◦
◦
◦
1
9
.5 K2O-10Cr2O3/Al2O3(T) calcined at (b) 500 C (c) 600 C (d) 700 C (e) 800 C (f)
00 C.
◦
the increasing reaction time, which was due to the coke formation
over the catalyst [3,9,14,46]. It was not difficult to notice that the
isobutane conversion and isobutene yield of the catalyst with lower
Cr2O3 loading decreased more slightly than that of the catalyst with
higher Cr2O3 loading during the dehydrogenation reaction, demon-
strating that the catalyst 5Cr2O3/Al2O3(600) exhibited much higher
stability among the series catalysts over the dehydrogenation.
The catalytic reactivity over the series catalysts 1.5 K2O-
dichromate/polychromate [24]. An additional weak band at 581 nm
observed for the series catalysts was probably due to another d-d
transition (A2g→T ) of Cr3+ in octahedral symmetry. The inten-
2g
sity of the latter band in the spectra increased with the increasing
3+
calcination temperature. Consequently, the amount of surface Cr
species steadily increased with the higher calcination temperature,
which was well consistent with XPS results above.
1
0Cr O /Al O (T) for the isobutane dehydrogenation are shown
2 3 2 3
in Fig. 6. By comparison with the catalyst 10Cr O /Al O (600), the
3
.2.6. H -TPR profiles
2
3
2
3
2
isobutane conversion had no significant increase over the K-doped
catalyst 1.5 K O-10Cr O /Al O (600). Nevertheless, the isobutene
The H -TPR profiles of the studied catalysts are described in
2
Fig. 4. All the samples showed a single H2 consumption peak
2
2
3
2
3
◦
selectivity greatly increased by 7.1–8.1%, indicating that potassium
was an effective promoter to the catalytic selectivity. Moreover, it
was observed that the isobutane conversion and isobutene selec-
tivity of the K-doped catalyst decreased more slightly than that of
(
Tmax) at 350–366 C, corresponding to the reduction of the
6
+
3+
dispersed Cr to Cr species [20,24,38,44]. The Tmax of the cat-
alyst 10Cr O /Al O (600) was located at 356 C. By comparison,
the similar Tmax 357 C was observed over the catalyst 1.5 K O-
1
was independent of the K O addition. It was noteworthy that,
moreover, the Tmax of the series catalyst 1.5 K O-10Cr O /Al O (T)
◦
2
3
2
3
◦
2
the catalyst 10Cr O /Al O (600), indicating that the deactivation
0Cr O /Al O (600), indicating that the reducibility of the catalyst
2
3
2
3
2
3
2
3
rate of the catalyst was greatly suppressed by K O addition. The
2
2
similar effect of K O over the dehydrogenation was reported by
2
2
2
3
2
3
the previous literatures [6,14,16,20,21].
was a slightly downward shift with the increasing calcination
temperature, suggesting that the interaction between Cr6+ and alu-
As for the influence of calcination temperature over the dehy-
drogenation reactivity (Fig. 6), the initial isobutane conversion
and initial isobutene yield increased to the maximum 60.1%
minium or oxygen species became weaker at the higher calcination
temperature.
3.3. Isobutane dehydrogenation
The products observed for isobutane dehydrogenation were
mainly isobutene and trace amounts of carbon monoxide, methane,
ethane, ethylene, propane, propylene and n-butene besides the
unreacted isobutane.
3.3.1. The dehydrogenation reactivity over the catalysts
The reactivity over the series catalysts xCr O /Al O (600) with
2
3
2
3
the increasing TOS for the isobutane dehydrogenation is plotted
in Fig. 5. As the Cr O3 loading increased, the initial isobutane
2
conversion and initial isobutene yield increased to the maxi-
mum 49.1% and 40.0% over the catalyst 10Cr O /Al O (600), then
2
3
2
3
slightly decreased thereafter. Meantime, the initial catalytic activity
−1
−1
(Table 2), demonstrat-
achieved at its maximum 18.3 mmol g
ing that the catalyst 10Cr O /Al O (600) possessed the highest
h
2
3
2
3
activity among the series catalysts [45]. It was observed that
the highest initial isobutene selectivity 84.9% was obtained over
the catalyst 20Cr O /Al O (600). For each catalyst, moreover, the
2
3
2
3
Fig. 6. The reactivity of the isobutane dehydrogenation over the series catalysts
isobutane conversion and isobutene yield gradually reduced with
1.5 K2O-10Cr2O3/Al2O3(T).

1
94
H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
and 56.1% over the catalyst 1.5 K O-10Cr O /Al O (800). Simulta-
2
2
3
2
3
neously, the catalyst exhibited the optimum initial catalytic activity
−
1
−1
2
2.4 mmol g
h among the series catalysts. It was observed
that the initial isobutene selectivity was up to 93.2% over the
◦
catalyst calcined at 800 C. With the increasing calcination tem-
perature, the initial isobutene selectivity increased and achieved at
the maximum 94.3% over the catalyst 1.5 K O-10Cr O /Al O (900).
2
2
3
2
3
Moreover, it was obviously noticed that the isobutane conversion
and isobutene yield over the catalyst calcined at higher temper-
ature decreased more quickly during the reaction, indicating that
the catalyst calcined at lower temperature was more stable over
the dehydrogenation. Also, the decreased isobutane conversion and
isobutene yield of each catalyst with the increase of the TOS over
the dehydrogenation was assigned to the coke deposition over the
catalyst [3,9,14,46].
3.3.2. The active chromium species
In case of the valence of the active chromium species, the active
3+
3+
2+
species in dehydrogenation is Cr , or both Cr and Cr , or coor-
Fig. 7. The ten dehydrogenation-regeneration cycles of the isobutane dehydrogena-
tion over the catalyst 1.5 K2O-10Cr2O3/Al2O3(800).
3
+
dinative unsaturated Cr according to the previous literatures
1,4,5,7,14,47–49]. Therefore, it is still an open question in the lit-
[
3+
eratures. In our experiment, it was suggested that Cr was mainly
the active site for isobutane dehydrogenation and the isobutene
respectively. Moreover, it was calculated that the catalytic activ-
−1 −1
−1 −1
h
ity was 29.6 mmol g
in the tenth cycle, which was maintainable during the ten cycles
Table 2). The basically unchanged activity and selectivity during
h
in the first cycle and 29.0 mmol g
3+
6+
selectivity was depended on the surface Cr /Cr value over the
catalyst.
(
Table 2 clearly shows the catalytic activity of isobutane dehy-
drogenation over the catalysts. With regard to the series catalysts
the ten dehydrogenation-regeneration cycles indicated the high
regenerative ability and high stability of the catalyst.
xCr O /Al O (600), it was observed that the catalytic activity
2
3
2
3
3+
increased with the increasing surface Cr amount over the catalyst
with chromia loading up to 10 wt.%. The catalyst xCr O /Al O (600)
3.3.4. The comparison of the activity with the reference catalyst
2
3
2
3
The catalyst 1.5 K O-10Cr O /Al O (800) was compared with
2
2
3
2
3
with chromia loading more than 15 wt.% exhibited the lower cat-
alytic activity than the catalyst 10Cr O /Al O (600), which was
the catalyst 1.5 K O-10Cr O /Al O (800)-R prepared by impreg-
2
2
3
2
3
2
3
2
3
nation method over the isobutane dehydrogenation (Fig. 8). It
was observed that the reference catalyst exhibited a slightly
higher initial isobutane conversion and comparable isobutene
ascribed to the larger and lower dispersed chromia particles
over the these catalysts, consistent with the PXRD results. It was
observed that the isobutene selectivity increased with the increas-
selectivity (about 94.0%) by contrast with the catalyst 1.5 K O-
ing Cr3+/Cr6+ value over the series catalysts xCr O /Al O (600),
2
2
3
2
3
1
0Cr O /Al O (800). This was probably assigned to that the refer-
2 3 2 3
agreement with the XPS results. For the series K-doped catalysts
3+
ence catalyst presented more surface Cr species and comparable
Cr /Cr
calcined at different temperatures, the catalytic activity increased
3+
6+
value with the catalyst 1.5 K O-10Cr O /Al O (800),
with the increasing surface Cr3+ amount over the catalysts. The
2 2 3 2 3
in accordance with the XPS characterization. However, the
isobutane conversion and isobutene yield of the reference
catalyst decreased more rapidly compared with the catalyst
slightly decreased activity over the K-doped catalyst calcined at
◦
9
00 C was probably assigned to the appearance of ␣-Al O in the
2
3
support and low specific surface area of the catalyst. Moreover, the
1
.5 K O-10Cr O /Al O (800), indicating that the catalyst 1.5 K O-
3+
6+
2 2 3 2 3 2
isobutene selectivity increased with increasing Cr /Cr value over
the series catalysts.
1
0Cr O /Al O (800) possessed a more stable activity during the
2
3
2
3
dehydrogenation. It may be due to the stabilization of the
Based on these results and the data in Table 2, we concluded that
the catalytic activity was improved with the increase of surface Cr3+
amount over the catalyst with highly dispersed chromium species
and the isobutene selectivity was enhanced with the increasing
3
+
6+
surface Cr /Cr value. In addition, the TOF (turnover frequency)
over the catalysts with highly dispersed chromium species were
−
1
3+
1
.6 ± 0.2 h . This meant that the activity per Cr site over the cat-
alysts was almost the same. Analogously, De Rossi et al. found that
the activity per atom of chromium was the same over the Cr-based
catalysts with different chromia loadings [47]. Lugo and Lunsford
also observed the same turnover frequency over the unsupported
and supported chromium oxide catalysts [50].
3.3.3. The dehydrogenation-regeneration cycles of the catalyst
In order to examine the regenerative ability and stability
of the catalyst during the isobutane dehydrogenation, the ten
dehydrogenation-regeneration cycles over the catalyst 1.5 K O-
2
1
0Cr O /Al O (800) were investigated (Fig. 7). It was calculated
2 3 2 3
that the initial isobutane conversion, isobutene selectivity and
yield were 55.3%, 92.7% and 51.3% in the first cycle, respectively.
In the tenth cycle, by comparison, the initial isobutane conver-
sion, isobutene selectivity and yield were 54.2%, 93.4% and 50.6%,
Fig. 8. The reactivity of isobutane dehydrogenation over the catalyst (a) 1.5 K2O-
10Cr2O3/Al2O3(800) and (b) 1.5 K2O-10Cr2O3/Al2O3(800)-R.

H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
195
support structure of the inner Cr3+ species in the catalyst 1.5 K O-
host and chromium precursor. The reactivity of isobutane dehy-
drogenation was investigated over the catalysts. The pore texture,
structural phase, the reducibility and the surface concentrations
of Cr3+ and Cr species over the catalysts were depending on
2
1
3
3
0Cr O /Al O (800) at calcination [11,41].
2 3 2 3
6+
.4. The characterizations of spent and regenerated catalysts
the chromia loadings and calcination temperature. The K O addi-
2
.4.1. N2 adsorption-desorption measurements
tion to the catalyst slightly decreased the specific surface area
and the surface Cr3+/Cr value. However, it greatly improved the
isobutene selectivity and reduced the deactivation rate. The cata-
6+
The BET specific surface areas and pore properties of the spent
and regenerated catalysts are presented in Table 1. Compared with
the fresh catalyst, the spent one had a slightly smaller specific sur-
◦
lyst 1.5 K2O-10Cr2O3/Al2O3(800) calcined at 800 C exhibited high
2
−1
3
−1
face area 190.9 m g , lower pore volume 0.65 cm g and smaller
pore size 12.28 nm. The reason for this was probably due to that
coke deposition on the spent catalyst inhibited a portion of the pore,
thus leading to the slight decrease in the textural performance of
the catalyst. During the ten dehydrogenation-regeneration cycles,
isobutene selectivity up to 93.2% at the highest isobutane conver-
sion of 60.1%. The catalytic activity increased with the increasing
amount of Cr3+ species over the catalysts with highly dispersed
chromium species and the catalytic selectivity was enhanced with
the increasing surface Cr3+/Cr value of the catalyst. The main-
tainable activity and selectivity over isobutane dehydrogenation
in the ten dehydrogenation-regeneration cycle demonstrated the
very high stability and regenerative ability of the catalyst. Compare
with the catalyst prepared by impregnation method, the catalyst
in this work presented more stable activity in the dehydrogena-
tion. This study presents a feasible way to facile synthesis of the
Cr2O3/Al2O3 catalysts with good catalytic performance and high
stability from a MOF material.
6+
2
−1
)
the basically invariant specific surface area (177.9 → 172.1 m g
3
−1
and pore volume (0.55 → 0.53 cm g ) of the regenerated catalyst
from the first to tenth cycle demonstrated its high stability of the
catalyst.
3.4.2. PXRD patterns
Fig. 2B displays the PXRD patterns of the spent and the tenth
regenerated catalyst 1.5 K O-10Cr O /Al O (800). It was observed
2
2
3
2
3
that no potassium and chromia phases appeared in the PXRD pat-
tern of the corresponding spent catalyst, implying that the reaction
conditions had no significant influence over the dispersion of these
species. The potassium and chromia phases were not visible over
the tenth regenerated catalyst, demonstrating the high phase sta-
bility of the catalysts during the ten cycles over the isobutane
dehydrogenation.
Acknowledgement
We gratefully acknowledge the financial support of the State
Key Laboratory for Oxo Synthesis and Selective Oxidation of China
and Suzhou Science and Technology Bureau of Applied Foundation
Research Project (SYG201219).
3
.4.3. TEM images
The TEM images of the spent catalysts are presented in Fig. 3g–i.
References
The worm-like mesopores of the spent catalysts suggested that the
pores were non-ordered, which were very similar to that of the
corresponding fresh catalysts. It was clearly seen that no chromia
particles were on the surface of the alumina support, indicating its
high dispersion over the spent catalysts. These results implied again
the high stability of the catalyst, which was in good agreement
with the N2 adsorption–desorption and PXRD characterizations.
Moreover, the morphology of the carbon residues was not clearly
observed on the spent catalysts, indicating the carbon residues
were amorphous.
[
[
[
1] S. De Rossi, G. Ferraris, S. Fremiotti, V. Indovina, A. Cimino, Appl. Catal. A: Gen.
106 (1993) 125–141.
2] F. Cavani, M. Koutyrev, F. Trifirò, A. Bartolini, D. Ghisletti, R. Iezzi, A. Santucci,
G.D. Piero, J. Catal. 158 (1996) 236–250.
3] S.M. Stagg, C.A. Querini, W.E. Alvarez, D.E. Resasco, J. Catal. 168 (1997) 75–94.
[4] B.M. Weckhuysen, A.A. Verberckmoes, J. Debaere, K. Ooms, I. Langhans, R.A.
Schoonheydt, J. Mol. Catal. A: Chem. 151 (2000) 115–131.
[
[
5] A.B. Gaspar, L.C. Dieguez, J. Catal. 220 (2003) 309–316.
6] M. Tasbihi, F. Feyzi, M.A. Amlashi, A.Z. Abdullah, A.R. Mohamed, Fuel Process.
Technol. 88 (2007) 883–889.
[7] A. Hakuli, A. Kytökivi, A.O.I. Krause, Appl. Catal. A: Gen. 190 (2000) 219–232.
[
8] S. Wang, K. Murata, T. Hayakawa, S. Hamakawa, K. Suzuki, Appl. Catal. A: Gen.
96 (2000) 1–8.
1
[
9] J. Gascón, C. Téllez, J. Herguido, M. Menéndez, Appl. Catal. A: Gen. 248 (2003)
105–116.
3
.4.4. XPS spectra
The XPS data of the spent and regenerated catalyst 1.5 K O-
[
[
10] R.G. Craig, J.M. Dufallo, Chem. Eng. Prog. 75 (1979) 62–65.
11] E.I. Nemykina, N.A. Pakhomov, V.V. Danilevich, V.A. Rogov, V.I. Zaikovskii, T.V.
Larina, V.V. Molchanov, Kinet. Catal. 51 (2010) 898–906.
2
1
0Cr O /Al O (800) are presented in Table 2. Compared with the
2 3 2 3
3+
fresh catalyst, it was observed that only Cr species were over
[12] S. Udomsak, R.G. Anthony, Ind. Eng. Chem. Res. 35 (1996) 47–53.
[13] S. De Rossi, M.P. Casaletto, G. Ferraris, A. Cimino, G. Minelli, Appl. Catal. A: Gen.
6
+
the spent catalyst, indicating that Cr species was reduced by
the reaction molecules. Moreover, the amounts of surface potas-
sium and chromium species both increased largely over the spent
catalyst, which was probably due to the decrease of the specific
surface area led to the low dispersion of the two species over
the surface of the catalyst. After re-calcination in air, the Cr
species were observed over the regenerated catalyst, suggesting
that a portion of Cr species were oxidized to Cr in the re-
167 (1998) 257–270.
[
14] B.M. Weckhuysen, R.A. Schoonheydt, Catal. Today 51 (1999) 223–232.
[15] S.T. Korhonen, S.M.K. Airaksinen, M.A. Ba n˜ ares, A.O.I. Krause, Appl. Catal. A:
Gen. 333 (2007) 30–41.
[
[
16] R.D. Cortright, J.A. Dumesic, J. Catal. 157 (1995) 576–583.
17] J.M. Hill, R.D. Cortright, J.A. Dumesic, Appl. Catal. A: Gen. 168 (1998) 9–21.
6
+
[18] Y.K. Park, F.H. Ribeiro, G.A. Somorjai, J. Catal. 178 (1998) 66–75.
[
19] B. Grzybowska- S´ wierkosz, Top. Catal. 21 (2002) 35–46.
3
+
6+
[20] E. Rombi, M.G. Cutrufello, V. Solinas, S. De Rossi, G. Ferraris, A. Pistone, Appl.
Catal. A: Gen. 251 (2003) 255–266.
[21] G.J. Siri, G.R. Bertolini, M.L. Casella, O.A. Ferretti, Mater. Lett. 59 (2005)
2319–2324.
3+
calcination process. The invariantly unchanged amount of Cr
3+
6+
species (1.41% → 1.40%) and Cr /Cr value (2.22 → 2.22) were
responsible for the maintainable activity and selectivity during the
ten dehydrogenation-regeneration cycles.
[
22] J. Słoczy n´ ski, B. Grzybowska, A. Kozłowska, K. Samson, R. Grabowski, A. Kotarba,
M. Hermanowska, Catal. Today 169 (2011) 29–35.
[23] S.M.K. Airaksinen, A.O.I. Krause, Ind. Eng. Chem. Res. 44 (2005) 3862–3868.
[24] D. Shee, A. Sayari, Appl. Catal. A: Gen. 389 (2010) 155–164.
[25] Y. Echegoyen, I. Suelves, M.J. Lázaro, M.L. Sanjuán, R. Moliner, Appl. Catal. A:
Gen. 333 (2007) 229–237.
4
. Conclusions
[
26] M.J. Lázaro, Y. Echegoyen, C. Alegre, I. Suelves, R. Moliner, J.M. Palacios, Int. J.
Hydrogen Energy 33 (2008) 3320–3329.
In summary, the no-ordered mesoporous Cr O /Al O cata-
2
3
2
2
3
−1
[27] M.A. Ermakova, D.Y. Ermakov, G.G. Kuvshinov, L.M. Plyasova, J. Catal. 187 (1999)
7–84.
lysts with large specific surface areas (149.4–381.6 m g ) and
7
3
−1
high pore volume (0.77–1.24 cm g ) were successfully synthe-
sized from a Cr-based MOF material MIL-101 as the molecular
[
28] G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I. Mar-
giolaki, Science 309 (2005) 2040–2042.

1
96
H. Zhao et al. / Applied Catalysis A: General 456 (2013) 188–196
[
[
[
29] O.I. Lebedev, F. Millange, C. Serre, G.V. Tendeloo, G. Férey, Chem. Mater. 17
[39] T.V.M. Rao, E.M. Zahidi, A. Sayari, J. Mol. Catal. A: Chem. 301 (2009)
159–165.
[40] G. Wang, L. Zhang, J. Deng, H. Dai, H. He, C.T. Au, Appl. Catal. A: Gen. 355 (2009)
192–201.
[41] N.V. Vernikovskaya, I.G. Savin, V.N. Kashkin, N.A. Pakhomov, A. Ermakova,
V.V. Molchanov, E.I. Nemykina, O.A. Parahin, Chem. Eng. J. 176-177 (2011)
158–164.
(
2005) 6525–6527.
30] M. Latroche, S. Surblé, C. Serre, C. Mellot-Draznieks, P.L. Llewellyn, J.H. Lee, J.S.
Chang, S.H. Jhung, G. Férey, Angew. Chem. Int. Ed. 45 (2006) 8227–8231.
31] P.L. Llewellyn, S. Bourrelly, C. Serre, A. Vimont, M. Daturi, L. Hamon, G.D.
Weireld, J.-S. Chang, D.-Y. Hong, Y.K. Hwang, S.H. Jhung, G. Férey, Langmuir
2
4 (2008) 7245–7250.
[
[
32] B. Liu, H. Shioyama, T. Akita, Q. Xu, J. Am. Chem. Soc. 130 (2008) 5390–5391.
33] R.K. Bhakta, J.L. Herberg, B. Jacobs, A. Highley, J.R. Behrens, N.W. Ockwig, J.A.
Greathouse, M.D. Allendorf, J. Am. Chem. Soc. 131 (2009) 13198–13199.
[42] S. Khaddar-Zine, A. Ghorbel, C. Naccache, J. Mol. Catal. A: Chem. 150 (1999)
223–231.
[43] M. Cherian, M.S. Rao, A.M. Hirt, I.E. Wachs, G. Deo, J. Catal. 211 (2002) 482–495.
[44] J.M. Kanervo, A.O.I. Krause, J. Catal. 207 (2002) 57–65.
[45] M. Boudart, Chem. Rev. 95 (1995) 661–666.
[34] W.X. Wang, Y.W. Li, R.J. Zhang, D.H. He, H.L. Liu, S.J. Liao, Catal. Commun. 12
(
2011) 875–879.
[
[
35] R. Das, P. Pachfule, R. Banerjee, P. Poddar, Nanoscale 4 (2012) 591–599.
36] A. Henschel, K. Gedrich, R. Kraehnert, S. Kaskel, Chem. Commun. (2008)
[46] T. Matsuda, I. Koike, N. Kubo, E. Kikuchi, Appl. Catal. A: Gen. 96 (1993) 3–13.
[47] S. De Rossi, G. Ferraris, S. Fremiotti, A. Cimino, V. Indovina, Appl. Catal. A: Gen.
81 (1992) 113–132.
4
192–4194.
[
[
37] H. Lu, D.H. Shen, C.L. Bao, Y.X. Wang, Phys. Status Solidi A 159 (1997) 425–437.
38] B. Grzybowska, J. Słoczy n´ ski, R. Grabowski, K. Wcisło, A. Kozłowska, J. Stoch, J.
Zieli n´ ski, J. Catal. 178 (1998) 687–700.
[48] A. Hakuli, M.E. Harlin, L.B. Backman, A.O.I. Krause, J. Catal. 184 (1999) 349–356.
[49] R. Puurunen, B.E. Weckhuysen, J. Catal. 210 (2002) 418–430.
[50] H.J. Lugo, J.H. Lunsford, J. Catal. 91 (1985) 155–166.