Potassium-modified molybdenum-containing SBA-15 catalysts for highly efficient production of acetaldehyde and ethylene by the selective oxidation of ethane

Article abstract of DOI:10.1016/j.jcat.2011.09.029

Mo-incorporated SBA-15 (Mo-SBA-15) catalysts with different Mo:Si molar ratios were synthesized by a direct hydrothermal method, and SBA-15-supported Mo catalysts (Mo/SBA-15 and K-Mo/SBA-15) were prepared by an incipient-wetness impregnation method. The structures of the catalysts were characterized by means of N₂ adsorption-desorption, XRD, TEM, FT-IR, and UV-Raman, and their catalytic performance for the oxidation of ethane was tested. Turnover frequency and product selectivity are strongly dependent on the molybdenum content and catalyst preparation method. Furthermore, the addition of potassium promotes the formation of isolated tetra-coordination molybdenum species and potassium molybdate. The K/Mo-SBA-15 catalysts possess much higher catalytic selectivity to acetaldehyde in the selective oxidation of ethane than the supported molybdenum catalysts (Mo/SBA-15 or K-Mo/SBA-15). The highest selectivity of CH₃CHO + C₂H₄ 68.3% is obtained over the K/Mo-SBA-15 catalyst. Analysis of kinetic results supports the conclusion that ethane oxidation is the first-order reaction and ethane activation may be the rate-determining step for the oxidation of ethane. Ethylene is a possible intermediate for acetaldehyde formation.

Full text of DOI:10.1016/j.jcat.2011.09.029

Journal of Catalysis 285 (2012) 134–144
Contents lists available at SciVerse ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Potassium-modified molybdenum-containing SBA-15 catalysts for highly
efficient production of acetaldehyde and ethylene by the selective oxidation
of ethane
a
a
a,
⇑
b,
⇑
b
a
a
Jian Liu , Lihong Yu , Zhen Zhao , Yongsheng Chen , Pengyu Zhu , Chao Wang , Yan Luo ,
a
a
a
Chunming Xu , Aijun Duan , Guiyuan Jiang
a
State Key Laboratory of Heavy Oil, College of Science, China University of Petroleum, Beijing 102249, People’s Republic of China
EMS Energy Institute and John and Willie Leone Family Department of Energy and Mineral Engineering, The Pennsylvania State University, University Park, PA 16802, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 April 2011
Revised 21 September 2011
Accepted 22 September 2011
Available online 29 November 2011
Mo-incorporated SBA-15 (Mo-SBA-15) catalysts with different Mo:Si molar ratios were synthesized by a
direct hydrothermal method, and SBA-15-supported Mo catalysts (Mo/SBA-15 and K-Mo/SBA-15) were
prepared by an incipient-wetness impregnation method. The structures of the catalysts were character-
2
ized by means of N adsorption–desorption, XRD, TEM, FT-IR, and UV-Raman, and their catalytic perfor-
mance for the oxidation of ethane was tested. Turnover frequency and product selectivity are strongly
dependent on the molybdenum content and catalyst preparation method. Furthermore, the addition of
potassium promotes the formation of isolated tetra-coordination molybdenum species and potassium
molybdate. The K/Mo-SBA-15 catalysts possess much higher catalytic selectivity to acetaldehyde in the
selective oxidation of ethane than the supported molybdenum catalysts (Mo/SBA-15 or K-Mo/SBA-15).
Keywords:
Selective oxidation of ethane
Molybdenum
Molybdenum-containing SBA-15
x
Highly dispersed and isolated MoO species
3 2 4
The highest selectivity of CH CHO + C H 68.3% is obtained over the K/Mo-SBA-15 catalyst. Analysis of
Catalysts
Kinetics
kinetic results supports the conclusion that ethane oxidation is the first-order reaction and ethane acti-
vation may be the rate-determining step for the oxidation of ethane. Ethylene is a possible intermediate
for acetaldehyde formation.
Ó 2011 Elsevier Inc. All rights reserved.
1
. Introduction
One of the major challenges for the petrochemical industry is to
at partial oxidation of ethane have been reported, viable processes
and effective catalysts have not been developed.
A variety of catalysts for the selective oxidation of ethane have
been examined. Vanadium- [5–7] and molybdenum- [3,8,9] based
catalysts have been widely studied in the selective oxidation of
alkanes. For example, Wang et al. found a new route for the selec-
tive oxidation of ethane to formaldehyde and acetaldehyde over
SBA-15-supported molybdenum catalysts [3], and Corberan et al.
prepared a novel V-UVM-7 bimodal mesoporous catalyst for the
selective oxidative activation of isobutane and obtained a high
selectivity of isobutene, 42.6% [10]. The performance of these cat-
alysts is strongly related to the nature of the active species. It has
been reported that isolated vanadium or molybdenum species with
low aggregation and/or coordination structure favor the selective
formation of oxygenates, whereas polymeric species or crystalline
manufacture valuable products by oxidizing lower alkanes instead
of olefins or aromatics for effective utilization of abundant natural
gas resources. The direct transformation of methane and ethane to
useful olefins and oxygenates has attracted much attention from
chemists and chemical engineers. Compared with the processes of
naphtha oxidation and methanol carbonylation (currently used to
obtain acetic acid), direct selective oxidation of ethane is more
attractive for easier separation of products and lower cost of ethane
[
1]. Usually, high temperatures (>500 °C) are required for the con-
version of ethane because of the low reactivity of its CAH bonds
2]. However, valuable oxygenates formed may be instantly con-
verted to CO and CO under such conditions. Thus, selective oxida-
[
2
tion of ethane to oxygenates is very difficult, and it is not an easy
task to obtain oxygenates with one-pass yields higher than 4%, even
if N O is used as a selective oxidant [2–4]. Although many attempts
2
x
MoO species are suitable for deep oxidation of alkanes [4,11,12].
In the past decades, highly dispersed catalysts with isolated active
sites have been developed, which may avoid overoxidation of the
partially oxidized products to undesired carbon oxides [4,7–12].
Moreover, the activity and the selectivity of the catalysts also seem
to depend on the nature of the support, and pure silica materials
with weak surface acidity are favorable for the oxidation of alkanes
to oxygenates. Mesoporous molecular sieves, MCM-41 and SBA-15,
⇑
Corresponding authors.
E-mail addresses: zhenzhao@cup.edu.cn (Z. Zhao), yzc2@psu.edu (Y. Chen).
0
021-9517/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2011.09.029

J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
Table 2
135
which have high surface area, large pore volume, regular structure,
Physical properties of Mo/SBA-15 and K-Mo/SBA-15 catalysts with different Mo
contents.
uniform pore size distribution, and relative high thermal stability,
are preferred supports for catalyst preparation [13,14]. However,
pure silica, SBA-15, shows very limited catalytic activity for its
weak redox properties, basicity, and acidity. Therefore, it is
necessary to introduce catalytic active sites into the framework
of SBA-15 and to improve its catalytic activity. For instance, the
incorporation of certain transition metal ions such as vanadium
into the framework of mesoporous molecular sieves can produce
catalysts with remarkable catalytic activity for many reactions
Sample
Mo:Si:molar ratio
S
BET, m2
g
V
P
, cm3
g
D, nm
I0.1
I2.5
I5
I8
I10
0.0025:100
0.038:100
0.045:100
0.13:100
1.85:100
6.79:100
0.0025:100
0.038:100
0.045:100
0.13:100
1.85:100
6.79:100
497.8
476.9
466.4
442.9
383.5
355.4
0.87
0.85
0.84
0.78
0.75
0.65
6.99
7.16
7.18
7.05
7.82
7.37
I12
IK0.1
IK2.5
IK5
IK8
IK10
IK12
396.7
397.7
398.4
331.0
135.4
0.75
0.75
0.73
0.62
0.51
7.60
7.59
7.35
7.51
15.21
[
3,15]. The catalysts, which have dispersed active species and in
which the active metals exist in a tetrahedral coordination struc-
ture, yield high selectivity to aldehyde. This kind of framework-
incorporated catalysts is different from supported catalysts. The
active metal ions are fully isolated and dispersed throughout the
support, and therefore, high concentrations of active sites can be
achieved in framework-incorporated catalysts. In contrast, in sup-
ported catalysts, the active phases are only present on the surface
or in the pores of the support and may aggregate further. The use of
vanadium incorporation into mesoporous silicon materials for the
activation of lower alkanes with oxygen was recently reported
Notes: F is denoted as framework-incorporation sample Mo-SBA-15; I is denoted as
incipient-wetness impregnation sample Mo/SBA-15; FK is denoted as the K/Mo-
SBA-15 sample after potassium was impregnated on Mo-SBA-15; IK is denoted as
the K-Mo/SBA-15 sample after potassium was further impregnated on Mo/SBA-15.
2. Experimental
[
10,15], but molybdenum incorporation into mesoporous frame-
works for the selective oxidation of lower alkanes has not been re-
ported so far.
2.1. Catalyst preparation
Alkali metals are another important component in the catalysts.
In most cases, they are employed as promoters to improve the cat-
alytic performance. In recent years, significant improvements in
the catalytic performance of alkali metal-modified catalysts have
been reported for several very important reactions [12,16–18].
The effect of alkali metal on the physicochemical characteristics
of catalysts commonly involves acid–base interaction, adsorp-
tion/desorption behavior, and the reducibility of the catalysts.
The addition of alkali metals can reduce the acidity of the catalyst
and increase the dispersion of the active sites [18]. All together,
these functions can inhibit the excessive oxidation of oxygenate
products and enhance the formation of aldehyde. Thus, the alkali
metal-modified Mo-incorporated mesoporous sieves may be good
catalysts for the selective oxidation of low alkanes.
Mo-incorporated mesoporous molecular sieves, Mo-SBA-15,
were synthesized by the direct hydrothermal (DHT) method
using tetraethyl orthosilicate (TEOS) and ammonium molybdate
((NH
4
)
6
7
Mo O
2
24ꢀ4H O) as starting materials. Nonionic triblock
copolymer surfactant, EO20PO70EO20 (P123, Aldrich, analytic grade),
was used as the structure-directing agent. The typical preparation
procedure of Mo-SBA-15 materials is described as follows. A quan-
tity of 8 g of P123 was dissolved in 240 mL HCl to form a solution(de-
noted as ‘‘solution A’’). A certain amount of ammonium molybdate
(the initial molar ratio of Mo to Si was 0.001, 0.025, 0.05, 0.08,
0.10, or 0.12) was dissolved in 60 mL of deionized water, and the
solution was stirred at room temperature for about 10 min and then
added into solution A. The resulting solution was heated to 40 °C and
kept at that temperature for 4 h while being stirred. Subsequently,
17 g TEOS was gradually added to the solution and continuouslystir-
red for 4 h. A small amount of aqueous ammonia was then added to
the solution to adjust the pH value to 3. The solution was stirred at
40 °C for 24 h. The resulting gel was placed in a Teflon autoclave
and maintained at 100 °C for 48 h. The solid products were filtered,
washed, and dried at 100 °C in air for 12 h and calcined at 600 °C for
4 h. In the filtration and washing processes, any Mo species that is
not incorporated into the catalysts would be washed away. Finally,
the mesoporous Mo-SBA-15 samples were obtained. The final Mo
In this work, we report the synthesis and catalytic behavior of a
novel framework-incorporated Mo-containing SBA-15 mesoporous
mixed oxide, denoted as Mo-SBA-15, for the selective oxidation of
ethane. The effects of the preparation method and the pres-
x
ence/absence of potassium on the structure of MoO in the Mo-
containing SBA-15 catalysts are investigated, and their catalytic
performance for the oxidation of ethane is compared. The kinetic
analysis and reaction pathways on the best K/Mo-SBA-15 catalysts
are also preliminarily discussed.
Table 1
Physical properties of Mo-SBA-15 and K/Mo-SBA-15 catalysts with different Mo contents.
Sample
Mo:Si (in recipe)/molar ratio
Mo:Si(by XRF):molar ratio
2h, ° (100)
d(100)
S
BET, m2
g
V
P
, cm3
g
D, nm
SBA-15
F0.1
F2.5
F5
F8
F10
0.837
0.797
0.740
0.820
0.819
0.800
–
0.900
0.919
0.879
0.920
0.880
10.6
11.1
11.9
10.8
10.7
11.0
–
9.8
9.6
11.0
9.6
10.0
849.5
915.1
872.4
776.9
837.5
640.9
382.9
415.3
403.2
375.9
325.4
285.5
52.3
1.21
1.26
1.27
1.10
1.15
0.93
0.71
0.77
0.72
0.67
0.57
0.46
0.16
5.68
5.52
5.84
5.66
5.47
5.80
7.37
7.40
7.12
7.08
6.85
6.48
11.90
0.1:100
2.5:100
5:100
8:100
10:100
12:100
0.5:0.1:100
0.5:2.5:100
0.5:5:100
0.5:8:100
0.5:10:100
0.5:12:100
0.0025:100
0.038:100
0.045:100
0.13:100
1.85:100
F12
6.79:100
FK0.1
FK2.5
FK5
FK8
FK10
FK12
0.5:0.0025:100
0.5:0.038:100
0.5:0.045:100
0.5:0.13:100
0.5:1.85:100
0.5:6.79:100
Notes: F is denoted as framework-incorporation sample Mo-SBA-15; I is denoted as incipient-wetness impregnation sample Mo/SBA-15; FK is denoted as the K/Mo-SBA-15
sample after potassium was impregnated on Mo-SBA-15; IK is denoted as the K-Mo/SBA-15 sample after potassium was further impregnated on Mo/SBA-15.

1
36
J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
contents in Mo-SBA-15 samples were determined by X-ray fluores-
cence (XRF) measurement.
In addition, supported molybdenum and potassium–molybde-
num catalysts including Mo/SBA-15, K-Mo/SBA-15, and K/Mo-
SBA-15 samples were prepared by the incipient-wetness impregna-
Mo-SBA-15, Mo:Si=X:100
SBA-15
(A)
F0.1
F2.5
tion method using ammonium molybdate and/or KNO
solution as Mo and/or K precursors. The amount of (NH
24ꢀ4H O was varied in order to achieve different Mo-to-Si ratios
3
aqueous
4 6
) Mo
F5
F8
7
O
2
ranging from 0.001 to 0.12, and whenever potassium was added,
the molar ratio between K and Si was kept at 0.005. The samples
were dried at 100 °C for 6 h and then calcined at 600 °C for 4 h in
air. In this work, all chemical reagents except P123 were purchased
from Beijing Chemical Reagent Company in China and were of ana-
lytic grade. For the sake of brevity, the catalysts are named F, I, FK,
and IK in the tables and figures. F denotes framework-incorporated
sample Mo-SBA-15; I denotes incipient-wetness impregnation sam-
ple Mo/SBA-15; FK denotes the K/Mo-SBA-15 sample after potas-
sium was impregnated on Mo-SBA-15; and IK denotes the K-Mo/
SBA-15 sample after potassium was further impregnated on Mo/
SBA-15.
F10
F12
1
2
3
4
5
2
Theta/degree
K/Mo-SBA-15, K:Mo:Si=0.5:X:100
(B)
SBA-15
FK0.1
2.2. Catalyst characterization
FK2.5
Nitrogen adsorption/desorption isotherms at ꢁ196 °C were re-
FK5
FK8
corded using a Micromeritics ASAP 2010 porosimeter. The samples
2
were outgassed in N flow at 80 °C for 4 h before the measure-
FK10
FK12
ments. X-ray fluorescence (XRF) measurements were taken on a
CSX-1000E spectrometer. Low-angle X-ray diffraction (XRD) pat-
terns were collected on a SIEMENS D5000 diffractometer in a 2h
range from 0.7° to 10° with a step size of 0.02°. Transmission elec-
tron microscopy (TEM) images were taken on a JEM-200CX instru-
ment at an acceleration voltage of 200 kV. The TEM samples were
sonicated and well suspended in ethanol. Drops of the suspension
were applied, and after drying, the fine particles were well dis-
persed on a copper grid coated with carbon. Fourier transform
infrared spectra (FT-IR) were collected on a FTS-3000 IR instru-
1
2
3
4
5
2
Theta/degree
Fig. 1. Low-angle powder XRD patterns of Mo-SBA-15 and K/Mo-SBA-15 samples
with different Mo:Si molar ratios.
ꢁ1
matography (GC) (SP3420, Beijing, People’s Republic of China)
equipped with two flame ionization detectors (FID) and two meth-
ane converters operating at 380 °C using a Ni catalyst. All lines and
valves between the exit of the reactor and the GC were kept at
20 °C to prevent condensation of the products. C
ganic oxygenates in the products, including CH CHO and acrolein,
were separated in a Porapak Q column. Separation of other compo-
ment running at 2 cm resolution. The samples were mixed with
KBr (sample:KBr = 1:100 mass ratio) and pressed into pellets. The
ꢁ
1
spectra were collected in the range of 4000 to 400 cm , and KBr
was used as background. UV-Raman spectra were obtained on a
LabRAM HR spectrometer (Horiba Jobin Yvon). The laser wave-
length was 325 nm. A laser power of 0.2 mW was applied at the
sample.
1
2 6 2 4
H , C H , and or-
3
nents, such as HCHO, CH
4 2
, CO, and CO , was carried out in a TDX-01
column.
2.3. Catalytic activity measurements
The catalytic reactions were carried out in a fixed-bed flow
reactor at atmospheric pressure. A quantity of 300 mg of the cata-
lyst was charged into the reactor. The catalyst particle size was in
the range 0.4–0.5 mm. The catalyst was pretreated in the quartz
reactor (6 mm i.d., 400 mm length) with a gas flow containing
3. Results and discussion
3.1. Structural characterization of the materials
high-purity N
with high-purity N
mixture, consisting of ethane and oxygen (C
passed through the catalyst bed (maintained at 575 °C) at a flow
rate of 15 ml min . All reactant gases were purchased from the
Dalian Date Standard Gas Company in China, and the purity was
greater than 99.999%. To accurately detect the true reaction tem-
perature, the catalyst bed was sandwiched between two quartz-
wool layers, and the tip of a K-type thermocouple was inserted into
the catalyst bed. The reaction temperature was controlled through
a PID-regulation system based on the measurements of an external
K-type thermocouple. The height of the catalyst bed was about
2
and O
at the same temperature. The reactant gas
/O = 3:1), was
2
at 575 °C for 30 min, followed by purging
The Mo content of Mo-SBA-15 and K/Mo-SBA-15 catalysts was
determined by XRF and is presented in Table 1. The final Mo con-
tent in each catalyst is very different from what was used in the
recipe, indicating that a large amount of Mo was not incorporated
into the framework of SBA-15. Therefore, when preparing the SBA-
15 supported catalysts (Mo/SBA-15 and K-Mo/SBA-15 samples),
we used the corresponding XRF-determined Mo contents in the
Mo-SBA-15 samples. Physical properties of the incorporated and
supported catalysts are reported in Tables 1 and 2, respectively.
For the samples with the highest Mo loading (nominal Mo content
x = 12), relatively more Mo remains in the catalysts. This might be
due to the collapse of the mesoporous structure of SBA-15 and the
2
H
2 6
2
ꢁ1
3
0 mm, and the length of the isothermal zone of the reactor oven
formation of nanoparticle MoO
the growth of crystal MoO species inside the mesoporous chan-
nels destroys the silica wall of SBA-15 [3].
x
species on these catalysts. Thus,
was 60 mm. The reaction space velocity (GHSV) was about
3
x
ꢁ1
200 h . The reaction products were analyzed by online gas chro-

J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
137
The XRD patterns in the range 0.5° < 2h < 5° of the Mo-SBA-15
and K/Mo-SBA-15 samples with different Mo:Si molar ratios are
shown in Fig. 1. The DHT samples with various Mo contents exhibit
a sharp peak at about 0.8° and two weak peaks at 1.4° and 1.6°, cor-
responding to the (100), (110), and (200) planes in SBA-15, which
are the characteristics of mesoporous material with a space group
of P6 mm symmetry [13]. This confirms the formation of the highly
ordered 2D-hexagonal P6 mm structure. As Mo content increases
to x = 12, these characteristic peaks almost disappear whether
potassium is present or not, indicating a decrease in structural reg-
ularity. Moreover, as shown in Table 1, the d spacing between
seems to reduce the (100) d spacing in all samples except FK5. This
clearly shows that the introduction of potassium affects the struc-
tures of the catalysts.
To gain more insight into the structure of DHT catalysts, mor-
phologies of the Mo-SBA-15 and K/Mo-SBA-15 catalysts were
examined by TEM. Fig. 2 shows TEM micrographs of typical Mo-
SBA-15 and K/Mo-SBA-15 samples. When the molar ratio of Mo
to Si is equal to 0.1 or 10, the TEM images show well-ordered hex-
agonal arrays of mesopores with one-dimensional channels, man-
ifesting a P6 mm mesoporous structure. The hexagonal arrays of
mesopores with one-dimensional channels are partially main-
tained in the F12 catalyst. However, the order of the mesoporous
structure has been destroyed. The addition of potassium causes
the partial collapse of the mesoporous structure in FK12.
Nitrogen adsorption–desorption isotherms for the catalysts
synthesized by different methods are presented in Fig. 3, and their
(
100) planes in the Mo-loaded catalysts is larger than that in the
pure siliceous SBA-15, suggesting the incorporation of Mo into
the framework of SBA-15. This is probably because the partial sub-
stitution of the larger Mo ion for Si results in the expansion of the
crystal structure of SBA-15. Moreover, the addition of potassium
Fig. 2. TEM images of Mo-SBA-15 and K/Mo-SBA-15. (a) Mo-SBA-15, Mo:Si = 0.1:100; (b) K/Mo-SBA-15, K:Mo:Si = 0.5:0.1:100; (c) Mo-SBA-15, Mo:Si = 10:100; (d) K/Mo-SBA-
5, K:Mo:Si = 0.5:10:100; (e) Mo-SBA-15, Mo:Si = 12:100; (f) K/Mo-SBA-15, K:Mo:Si = 0.5:12:100.
1

1
38
J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
K/Mo-SBA-15
Mo-SBA-15, Mo:Si=X:100
K:Mo:Si=0.5:X:100
(
A)
(C)
FK0.1
F0.1
F2.5
F5
FK2.5
FK5
FK8
F8
FK10
FK12
F10
F12
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure/P/PO
Relative pressure/P/PO
Mo/SBA-15, Mo:Si=X:100
K-Mo/SBA-15,
K:Mo:Si=0.5:X:100
(
B)
(D)
I0.1
I2.5
IK0.1
IK2.5
IK8
I8
I10
IK10
IK12
I12
0.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure/P/PO
Relative pressure/P/PO
Fig. 3. Nitrogen adsorption–desorption isotherms of catalysts with different Mo:Si molar ratios. (A) Mo-SBA-15; (B) Mo/SBA-15; (C) K/Mo-SBA-15; (D) K-Mo/SBA-15.
ꢁ
1
physical properties are presented in Tables 1 and 2. The DHT and
IWI catalysts have similar textural properties. The isotherms of
the catalysts with low Mo content are of type IV with an H1 hyster-
esis loop, showing typical characteristics for mesoporous materials
with a 2D-hexagonal structure [13]. A general trend of decreasing
pore volume is observed with increasing Mo loading. A dramatic
decrease in pore volume with high Mo loading, as in FK12, is
noticed. Moreover, the surface area of the K-modified sample
is remarkably lower than that of the corresponding Mo-SBA-15
catalyst when the Mo content is the same. This may be due to
the introduction of K blocking up the mesoporous channel of
SBA-15. The introduction of K plays a major role in the K/Mo-
SBA-15 catalysts; their surface areas, in fact, look closer to those
of the catalysts prepared by incipient-wetness impregnation. How-
ever, Mo is incorporated into the framework of SBA-15 and does
not occupy the pore channels of SBA-15 for the Mo-SBA-15 cata-
lysts. Therefore, the introduction of Mo only slightly affects the
surface areas of catalysts. Thus, the difference in the surface area
of catalysts doped with K and Mo may suggest that Mo should
be incorporated into the framework of SBA-15.
band at 953 cm is not detected for the IWI Mo/SBA-15 samples,
as shown in Fig. 4B, suggesting that the molybdenum atom cannot
be incorporated into the framework of SBA-15 by the impregnation
method. It is also not observed in Fig. 4C for K/Mo-SBA-15 samples,
which might be due to the addition of potassium covering the IR
vibration peak of the Si–O–Mo bond.
UV-Raman spectroscopy is a powerful technique for the charac-
terization of catalytic materials and has been widely used to char-
acterize molecular sieves. Fig. 5 shows the UV-Raman spectra of
the DHT Mo-SBA-15 and the IWI Mo/SBA-15 catalysts. For compar-
ison, the Raman spectrum of SBA-15 is also included in Fig. 5A. The
Raman bands at 500, 604, 810, and 978 cm are observed on all
samples. The bands at 500 and 604 cm are assigned to the three-
and four-silane rings, and the band at 810 cm is ascribed to the
siloxane linkage. The band at 978 cm is associated with defect
sites such as the surface silanol groups [3,24,25]. For the sample
with a very low Mo content (x = 0.1), a weak band at 969 cm is
detected. With the increase in Mo content from x = 0.1 to x = 12,
a shoulder band at 980–990 cm is also discernable. The same
phenomenon has been reported for the MoO /SiO catalysts under
dehydration [3,24]. The band around 980 cm can be assigned to
the stretching vibration of terminal Mo@O of highly dispersed
ꢁ
1
ꢁ
1
ꢁ
1
ꢁ1
ꢁ1
ꢁ1
x
ꢁ
2
1
The FT-IR spectra of the DHT and the IWI catalysts are shown in
ꢁ1
Fig. 4. Two characteristic bands at ꢂ463 and ꢂ802 cm can be
ꢁ1
detected in all samples. The band centered at 463 cm is assigned
to the flexural vibrations of Si–O–Si, and the band centered at
MoO
band at 969 cm may be ascribed to a highly dispersed surface
molybdenum species with interaction with SiO [3,26]. The
x x
species, such as monomeric MoO moieties, whereas the
ꢁ
1
ꢁ1
about 802 cm is ascribed to the symmetric stretching vibration
2
ꢁ1
of tetrahedral SiO
be assigned to the stretching vibration mode of an [SiO
bonded to a heteroatom (Si–O–M) in silicate molecular sieves
23], suggesting the incorporation of Mo into the framework of
molecular sieves in the DHT Mo-SBA-15 catalysts. However, this
4
[19–22]
.
In Fig. 4A, the band at 953 cm can
absence of Mo–O–Mo stretching in these samples indicates that
the DHT catalysts with Mo content x 6 12 mainly contain highly
x
dispersed MoO species. The UV-Raman spectra of Mo/SBA-15
samples are shown in Fig. 5B. For the sample with very low Mo
content (x = 0.1), similar to the DHT catalysts, only the stretching
4
] unit
[

J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
139
Mo-SBA-15, Mo:Si=x:100
969
(
A)
Mo-SBA-15,
Mo:Si=X:100
9
90
025
(A)
F0.1
F1
1
F2.5
F5
F12
F10
F12
F10
F8
802
953
4
63
F5
MoO3
F2.5
F0.1
4
00
600
800
1000
1200
1400
1400
1400
500
6
600
04
810
978
SBA-15
-
1
Wavenumber/cm
2
00
400
800
1000 1200 1400 1600
-
1
Raman shift/cm
Mo/SBA-15, Mo:Si=x:100
(
B)
9
60
8
10
1
030
4
60
Mo/SBA-15,
Mo:Si=x:100
9
76
(B)
I12
I10
8
76
8
20
I12
I8
8
40
I5
I2.5
I0.1
I10
I8
7
90
400
600
800
1000
1200
I2.5
I0.1
-
1
Wavenumber/cm
K/Mo-SBA-15, K:Mo:Si=0.5:x:100
(
C)
FK0.1
FK1
200
400
600
800
1000 1200 1400 1600
-1
Raman shift/cm
FK2.5
FK5
FK10
K/Mo-SBA-15,
K:Mo:Si=0.5:x:100
978
(C)
FK12
807
953
463
872
MoO3
820
FK12
400
600
800
1000
1200
-1
892
FK10
FK8
Wavenumber/cm
1
085
Fig. 4. FT-IR spectra of samples with different Mo contents. (A) Mo-SBA-15; (B) Mo/
SBA-15; (C) K/Mo-SBA-15.
FK5
FK2.5
x
mode of terminal Mo@O of isolated momomeric MoO species
FK0.1
ꢁ1
centered at 976 cm is observed, and no Mo–O–Mo stretching
vibration of polymeric Mo species is detected. However, with
2
00
400
600
800
1000 1200 1400 1600
ꢁ1
increasing Mo content, the bands at 820–880 cm corresponding
to Mo–O–Mo stretching vibration of polymeric Mo species are ob-
served even at low Mo content (x = 2.5). This means that Mo spe-
cies in the IWI catalysts have more tendencies to aggregate than
-1
Raman shift/cm
Fig. 5. Raman spectra of catalysts with different Mo contents. (A) Mo-SBA-15;
(B) Mo/SBA-15; (C) K/Mo-SBA-15.
ꢁ1
those in the DHT catalysts. As x P 10, the band at 976 cm as-
signed to Mo@O stretching vibration becomes dominant. The coex-
istence of Mo@O and MoAOAMo bonds indicates that crystalline
Fig. 5C shows the UV-Raman spectra of K/Mo-SBA-15 samples.
For the sample with very low Mo content (x = 0.1), no obvious
difference is observed between K/Mo-SBA-15 and Mo-SBA-15.
Besides the weak band at 969 cm , another weak band at
980–990 cm can also be discerned. When Mo content increases
to x = 2.5 and x = 5, two new Raman peaks at 892 and 1085 cm
emerge, indicating the formation of potassium molybdate, and
these two Raman peaks become weaker with further increase in
MoO
x P 10). Thus, UV-Raman results further indicate that the highly
dispersed and isolated MoO species in high concentration were
3
species are formed in the IWI catalysts at high Mo content
ꢁ1
(
ꢁ1
x
ꢁ1
obtained in the Mo-SBA-15 catalysts by the framework-incorpora-
tion method but not in the Mo/SBA-15 catalysts by the incipient-
wetness method.

1
40
J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
K
2
MoO
4
[28]. Therefore, the observed changes in the Raman spec-
interac-
(
A)
5
4
3
2
1
0
0
0
0
0
tra of Fig. 5C can also be interpreted in terms of the MoO
tion with potassium to form potassium metal molybdates. The
sharp peaks of the K/Mo-SBA-15 (K/Mo > 1, mol) at 892 and
K/Mo-SBA-15
Mo-SBA-15
Mo/SBA-15
x
K-Mo/SBA-15
ꢁ1
1
085 cm may be ascribed to the formation of K
2
MoO
4
[20,29],
the isolated Mo species dominating the surface of the sample.
When Mo content increases to x = 10 (K/Mo < 1, mol), several
ꢁ1
new Raman peaks at 820, 872, and 978 cm are observed, indicat-
ing the formation of two-dimensional polymolybdates, which were
probably ascribed to K
the presence of the alkali metal K modified the structure of the
molybdenum oxide phase and formed new species of K Mo
and K MoO
2 2 7
Mo O [19,30]. All results have shown that
2
2 7
O
2
4
.
0
2
4
6
8
10
12
Mo content (K:Mo:Si=0.5:x:100)
3.2. Comparison of the catalytic performances of DHT and IWI
catalysts
Mo-SBA-15
K/Mo-SBA-15
(
B)
0
0
0
0
.03
.02
.01
.00
Fig. 6 displays the results of C
alysts at 575 °C. As shown in Fig. 6A, in the absence of potassium,
conversion gradually increases with increasing Mo content,
and it reaches a maximum at x = 10 for the DHT catalysts and
x = 8 for the IWI catalysts, respectively. The addition of potassium
has a complex effect on the ethane conversion. Table 3 also shows
that a decrease in C H conversion is observed for K/Mo-SBA-15
2 6
catalysts as Mo content changes from x = 0.1 to x = 2.5; it is
remarkably enhanced with increasing Mo content and reaches a
2 6
H conversion over different cat-
2 6
C H
maximum value of 50.4% at x = 8, and then C
creases slightly with a further increase in Mo content (x = 10 or
2). As shown in Fig. 6B, the ethane conversion normalized on unit
2 6
H conversion de-
0
2
4
6
8
10
12
1
Mo content (K:Mo:Si=0.5:x:100)
surface area is constant for all Mo-SBA-15 catalysts, whereas it
gradually increases with increasing Mo content for K/Mo-SBA-15
catalysts. A sharp increase in ethane conversion per unit surface
area at x = 12 may be attributed to the severe reduction in the spe-
cific surface area of this catalyst (see Table 1).
Fig. 6. (A) Ethane conversion over the four systems of Mo-SBA-15, Mo/SBA-15, K/
Mo-SBA-15, and K-Mo/SBA-15 and (B) ethane conversion normalized to unit surface
area over the Mo-SBA-15 and K/Mo-SBA-15 catalysts as a function of Mo content.
Fig. 7 displays the yield of ethylene and acetaldehyde and the
total yield of ethylene and acetaldehyde in the selective oxidation
of C H over different catalysts at 575 °C. In the absence of
Mo content. The characteristic peaks of different potassium molyb-
dates are at very similar or even the same frequencies [27], so it is
very difficult to distinguish them. Several studies [27–30] have
shown that some two-dimensional compounds without well-de-
fined structures may be formed at low alkali metal content and
at higher alkali metal content well-defined crystalline structures
are more likely to form. Verbruggen et al. [27] studied alkali
ion-modified silica-based Mo oxides and concluded that in the
uncalcined samples with K/Mo 6 1, many sixfold-coordinated Mo
2
6
potassium, the yield of ethylene on the DHT catalysts is lower than
that on the IWI catalysts when Mo content x is in the range
0.1 < x < 10 (Fig. 7A). When x > 10, the yield of ethylene on the
DHT catalyst is higher than that on the IWI catalyst. Furthermore,
it decreases with the further increase in Mo content no matter
whether it is on the DHT or on the IWI catalysts. The yield of acet-
aldehyde increases with the increasing Mo content at medium
molybdenum content, as shown in Fig. 7B. Further increase in
Mo content (x P 10) causes a decrease in the yield of acetaldehyde.
When Mo content is low, the yield of acetaldehyde on the catalysts
species are present; mainly, K
2
Mo
2
O
7
is detected. At K/Mo > 1.0,
MoO is
the Mo coordination changes into fourfold, and mainly K
2
4
detected, which is differently distorted than in crystalline
Table 3
The catalytic performance of K/Mo-SBA-15 catalysts in the selection oxidation of ethane at 575 °C.
Sample
K:Mo:Si:molar ratio
T, °C
C
H
2 6
Conv., %
Selectivity, %
Selectivity, %
a
Olefinb
Alkanec
CO
x
Aldehydes
Olefins + aldehydes
C1
C2
Totald
FK0.1
FK2.5
FK5
FK8
FK10
FK12
0.5:0.0025:100
0.5:0.038: 100
0.5:0.045: 100
0.5:0.13: 100
0.5:1.85: 100
0.5:6.79: 100
575
575
575
575
575
575
21.8
9.4
18.7
50.4
47.1
45.1
44.7
29.5
36.7
32.2
31.9
39.4
49.1
64.5
53.3
43.0
38.3
45.0
1.8
2.1
2.3
7.3
7.9
6.4
0.9
2.7
1.0
0.5
0.1
1.6
3.3
1.1
6.7
16.9
21.7
7.1
4.3
3.8
7.7
17.4
21.8
8.9
53.4
68.3
61.0
60.4
60.1
53.9
Notes: C1: formaldehyde, C2: acetaldehyde. F is denoted as framework-incorporation sample Mo-SBA-15; I is denoted as incipient-wetness impregnation sample Mo/SBA-15;
FK is denoted as the K/Mo-SBA-15 sample after potassium was impregnated on Mo-SBA-15; IK is denoted as the K-Mo/SBA-15 sample after potassium was further
impregnated on Mo/SBA-15.
a
x 2
CO : CO and CO .
Olefins: ethylene, propylene, and butylene.
Alkane: methane and propane.
b
c
d
Total: formaldehyde, acetaldehyde, and acrolein.

J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
141
is similar. When x is greater than 2.5, with the increased Mo con-
tent, the yield of acetaldehyde on the DHT catalysts is remarkably
higher than that on the IWI catalysts, indicating that isolated Mo
species favor the formation of acetaldehyde. The yield of ethylene
greatly increases after the addition of potassium to the Mo-SBA-15
catalysts, as shown in Fig. 7A. Under the same reaction conditions,
the maximum ethylene yield is only 8.9% for Mo-SBA-15 catalyst at
x = 8. However, it reaches 19.9% after the addition of potassium at
x = 8, and it is still very high at x = 10 (18.1%). In K/Mo-SBA-15
catalysts, the addition of potassium does not cause much enhance-
ment in the yield of acetaldehyde when Mo content is low (x = 0.1,
ilar. With increasing Mo content, the yield of acetaldehyde on the
DHT catalysts is much greater than that on the IWI catalysts. As
shown in Fig. 7B, the maximal yield of acetaldehyde is 10.2% for
K/Mo-SBA-15 catalyst at x = 10. As far as we know, this is the high-
est one-pass yield of acetaldehyde in the literature as yet for the
selective oxidation of ethane to produce aldehyde oxygenates.
Fig. 7C shows the total yield of acetaldehyde and ethylene over dif-
ferent systems of catalysts. The total yield is always greater on K/
Mo-SBA-15 catalysts than on other three systems of catalysts with
the same Mo content. Thus, it is very interesting to investigate the
reaction kinetics for the selective oxidation of ethane onto K/Mo-
SBA-15 catalysts.
2
.5, and 5). As Mo content increases further, the yield increases
remarkably after the addition of potassium. When Mo content is
up to x = 12, yield decreases, as shown in Fig. 7B. When Mo content
is low, the yield of acetaldehyde on the DHT or IWI catalysts is sim-
Turnover frequency (TOF) is defined as number of ethane con-
versions per Mo site per second. Ethane conversion over Mo sites
in a catalyst is determined by subtracting the conversion over pure
SBA-15 from the total conversion over the catalyst. To calculate
TOF values, the active sites should be identified and quantified;
however, this is almost impossible even if dispersion data are
available, because not all surface Mo atoms are equally active.
Therefore, we use the ‘‘apparent TOF’’ to assess an ‘‘averaged’’ effi-
ciency of Mo in different catalysts for catalyzing the reaction [31].
For incorporated/supported Mo oxide catalysts, the isolated/poly-
meric sites are known to be active sites and the number can be
estimated based on the molybdenum loading. Fig. 8 shows the
TOF as a function of the Mo content. A very high TOF is achieved
on all catalysts when the Mo content of x is equal to 0.1. The
decrease in TOF with increasing Mo content may be due to some
structural modifications of the surface molybdenum species. It
may indicate that the highly dispersed Mo on SBA-15 plays an
important role in the oxidative conversion of ethane to acetalde-
hyde and ethylene. The addition of potassium into catalysts
increases the TOF. The TOF value reaches a maximum at x = 8 for
K/Mo-SBA-15 catalysts due to the formation of different potassium
molybdates, further indicating that the selective oxidation of eth-
ane is a structure-sensitive reaction.
To investigate the stability of catalysts, the effects of reaction
time on ethane conversion and on product selectivity at 575 °C
over K/Mo-SBA-15 (nominal molar ratio K:Mo:Si = 0.5:10:100)
were also investigated. The results are depicted in Fig. 9. Only a
small change is observed for the product distribution with time
on stream. The conversion of ethane was maintained at 47%. At
the same time, there was only a slight effect of the reaction time
x
on the selectivity of ethylene, acetaldehyde, and Co , as shown in
Fig. 9A. These results indicate that the catalyst was stable under
the reaction conditions.
(
A)
2
1
1
0
5
0
5
0
K/Mo-SBA-15
Mo-SBA-15
Mo/SBA-15
K-Mo/SBA-15
0
0
0
2
4
6
8
10
12
12
12
Mo content (K:Mo:Si=0.5:x:100)
1
1
2
0
8
6
4
2
0
(
B)
K/Mo-SBA-15
Mo-SBA-15
Mo/SBA-15
K-Mo/SBA-15
2
4
6
8
10
Mo content (K:Mo:Si=0.5:x:100)
0
.55
0.50
.45
0.40
3
2
2
1
1
0
5
0
5
0
5
0
K/Mo-SBA-15
Mo-SBA-15
Mo/SBA-15
K/Mo-SBA-15
(
C)
0
Mo-SBA-15
Mo/SBA-15
K-Mo/SBA-15
K-Mo/SBA-15
0
0
0
0
0
0
0
0
.35
.30
.25
.20
.15
.10
.05
.00
2
4
6
8
10
0
2
4
6
8
10
12
Mo content (K:Mo:Si=0.5:x:100)
Mo content (Mo:Si=x:100)
Fig. 7. Yield of ethylene and acetaldehyde over the four systems of Mo-SBA-15, Mo/
Fig. 8. TOF of ethane conversion over Mo-SBA-15, Mo/SBA-15, K/Mo-SBA-15, and
SBA-15, K/Mo-SBA-15, and K-Mo/SBA-15 catalysts as a function of Mo content.
K-Mo/SBA-15 catalysts as a function of Mo content.

1
42
J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
5
4
3
2
1
0
0
0
0
0
0
K:Mo:Si=0.5:10:100
(
A)
0
.6
2
C H4
K/Mo-SBA-15
(
A)
COX
0.5
0
0
0
0
.4
.3
.2
.1
CH CHO
3
x=0.1
x=2.5
x=8
x=10
x=12
2
4
6
8
10
Time on stream/h
3
0
40
50
60
70
80
Ethane Content / %
4
3
3
2
2
1
1
0
5
0
5
0
5
0
5
55
K:Mo:Si=0.5:10:100
(B)
K/Mo-SBA-15
(
B)
0
.5
5
0
5
0
C H conversion
2
6
4
4
0.4
x=0.1
x=2.5
x=8
x=10
x=12
C H4
2
0.3
CH CHO
3
35
30
0.2
2
4
6
8
10
0.1
20
30
40
50
Time on stream/h
Oxygen Content / %
Fig. 9. Catalytic performance of K/Mo-SBA-15 catalysts (K:Mo::Si = 0.5:10:100) as a
function of reaction time for the selectivity oxidation of ethane. Reaction condi-
Fig. 11. Effect of ethane and oxygen content on ethane conversion rate over
tions: W = 0.3 g, T = 575 °C, F = 900 ml/min, C
2
H
6
/O
2
= 3:1.
different K/Mo-SBA-15 catalysts.
5
5
4
4
3
3
2
2
1
1
5
0
5
0
5
0
5
0
5
0
5
0
C
2
H
6
changes little above 550 °C, and it even decreases slightly at
600 °C. Therefore, the best optimal temperature is 575 °C for the
selective oxidation of ethane. The conversion of C reaches a
0
2
5
8
1
1
.1
.5
Mol ratio
K:Mo:Si=
2
H
6
0
.5:X:100
maximum of 50.2% when the reaction temperature is 575 °C at
x = 8.
0
2
3.3.2. Effect of reactant concentration on ethane oxidation
The rate of ethane consumption increases almost linearly with
the increase in ethane concentration, as illustrated in Fig. 11A.
However, the rate of C
concentration when oxygen content in reactant gases is 0.2–
.5, as can be observed in Fig. 11B. Kinetic data imply that the
2 6
H oxidation appears to be insensitive to
O
0
2
selective oxidation of ethane is approximately first order with re-
spect to ethane and is consistent with a mechanism in which the
breaking of a CAH bond is rate-determining. The reaction was
found to be almost independent of the oxygen concentration, prob-
ably indicating that under reaction conditions, the reoxidation of
reduced catalyst is not the rate-determining step, which is consis-
tent with other reported results [32–34].
4
80
500
520
540
560
580
600
o
Temperature/ C
Fig. 10. Conversion of ethane as a function of reaction temperature over K/Mo-SBA-
5 catalysts.
1
3
.3. Kinetic investigation on the catalytic oxidation of ethane on the
3.3.3. Effect of ethane or ethylene conversion on the selectivity of
products
framework-incorporated K/Mo-SBA-15
To elucidate the reaction network for ethane oxidation, the
experiments were performed using different hydrocarbons as reac-
tants, such as ethane or ethylene. The selectivity of products for the
partial oxidation of ethane as a function of ethane conversion is
illustrated in Fig. 12. As the ethane conversion increases, the selec-
3
.3.1. Effect of temperature
The conversion of C H as a function of temperature over K/Mo-
2
6
SBA-15 catalysts at constant space velocity is illustrated in Fig. 10.
For catalysts with different Mo contents, the conversion of ethane
increases with increasing reaction temperature. The conversion of
3
tivity of ethylene decreases, while the selectivity of CH CHO

J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
143
70
60
50
40
30
20
10
0
30
ethylene
K/Mo-SBA-15, x = 8 (A)
(
A)
2
2
1
1
5
0
5
0
5
0
ethane
ethylene
K/Mo-SBA-15 (K:Mo:Si = 0.5:8:100)
acetaldehyde
0
10
20
30
40
50
350
400
450
500
550
o
Conversion of ethane / %
Temperature / C
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
9
0
ethane
(
B)
(B)
80
ethylene
7
6
5
4
0
0
0
0
ethylene
K/Mo-SBA-15 (K:Mo:Si = 0.5:10:100)
acetaldehyde
K/Mo-SBA-15, x = 8
30
2
1
0
0
0
10
20
30
40
50
3
50
400
450
500
550
600
Conversion of ethane / %
o
Temperature / C
Fig. 12. Selectivity of ethylene and acetaldehyde as a function of ethane conversion
over the typical K/Mo-SBA-15 catalysts (x = 8, 10).
20
(
C)
ethane
ethylene
increases at the expense of C
2 4
H . This indicates that ethylene ap-
15
10
5
pears as the primary product of the partial oxidation of ethane.
The main useful products at 575 °C for K/Mo-SBA-15 catalyst (with
x = 8) are found to be ethylene and acetaldehyde, with selectivity
of 39.5% and 19.9%, respectively. The decrease in ethylene selectiv-
ity with corresponding increase in acetaldehyde selectivity indi-
cates that acetaldehyde is the secondary product of consecutive
oxidation of ethylene. The selectivity and yield of products for
the partial oxidation of ethylene or ethane over K/Mo-SBA-15 with
x = 8 catalyst as a function of temperature are illustrated in Fig. 13.
The selectivity and yield of acetaldehyde from ethylene are much
greater than those from ethane. These results may further show
that acetaldehyde originates directly from ethylene.
K/Mo-SBA-15, x = 8
0
350
400
450
500
550
o
Temperature / C
Fig. 13. (A) Conversion of ethylene and ethane (B) selectivity and (C) yield of
acetaldehyde for partial oxidation over the typical K/Mo-SBA-15 catalyst (x = 8) as a
function of reaction temperature.
3
.3.4. Reaction pathway for ethane oxidation
Study on the kinetics of selective oxidation of alkanes is helpful
in obtaining important information about the reaction mechanism.
Even though the kinetic studies cannot determine the molecular
mechanism, they allow the exclusion of certain reaction pathways
and help evaluate particular reaction routes and different catalytic
sites in the proposed mechanism [32,35–37]. Three possible routes
can be considered for the acetaldehyde formation: (1) direct oxida-
tion of ethane, (2) ethanol as an intermediate, and (3) ethylene as
an intermediate:
No ethanol was detected in the ethane oxidation reactions over
K/Mo-SBA-15 catalysts. Therefore, acetaldehyde formation via Eq.
(
2) can be excluded from possible reaction routes under these reac-
tion conditions. This may suggest that CH CHO should be produced
3
by the reaction pathways of Eqs. (1) and (3). The reactivity of eth-
ane is much lower than that of ethylene, and the yield of acetalde-
hyde from ethane oxidation is also much lower than that from
ethylene, as shown in Fig. 13. Therefore, it is not likely that acetal-
dehyde is mainly produced from the direct oxidation of ethane via
Eq. (1).
C
C
C
2
2
2
H
H
H
6
6
6
! CH
! CH
3
CHO
ð1Þ
ð2Þ
ð3Þ
3
CH
2
OH ! CH
3
CHO
The reaction scheme for ethane oxidation includes parallel reac-
tions forming ethylene and acetaldehyde, and ethylene may be fur-
! C
2
H
4
! CH CHO
3

1
44
J. Liu et al. / Journal of Catalysis 285 (2012) 134–144
ther oxidized to acetaldehyde. Linke et al. reported that the consec-
utive reactions are favored at low temperatures [38]. Our results
exhibit that ethylene as an intermediate in the consecutive
reaction is favored at relatively high temperatures. This may be be-
cause the addition of potassium, which is favorable for the desorp-
tion of ethylene and acetaldehyde, mitigates the deep oxidation of
ethane [12]. It is clear that the activation energy of the ethane oxi-
dation to ethylene must be lower than the activation energy of the
tive oxidation of ethane to acetaldehyde reported so far in the
literature.
Acknowledgments
This work was supported by the National Natural Science Foun-
dation of China (Nos. 20773163, 20833011, 20803093, 21173270,
and 21177160), the Program for New Century Excellent Talents
in the University of China (NCET-10-0811), the 863 Program of Chi-
na (No. 2009AA06Z313), the Beijing Outstanding Ph.D. Thesis
Foundation (No. YB 20091141401), a CNPC Project (2011D-4604-
0101), the Innovation Foundation (2010D-5006-0402), and the
Doctor Select Foundation (No. 200804251016).
x
direct oxidation of ethane to acetaldehyde. CO is formed from eth-
ane, ethylene, and acetaldehyde. In finding out which reaction is
the rate-determining step in the oxidation of ethane, the following
reasons are considered. Acetaldehyde desorption cannot be the
rate-determining step, since different rates of acetaldehyde forma-
tion were found in ethylene and ethane oxidation. The results re-
veal as well that catalyst reoxidation is not the rate-determining
step, since the oxygen content did not affect the ethane oxidation
rate in ethane conversion. In addition, the lower ethane oxidation
rate also suggests that the activation of the CAH activation in
ethane, not the reoxidation of the catalyst, should be the rate-con-
trolling step. This is because, if the latter applied, then similar rates
of oxidation should be expected for both ethane and ethylene.
Thus, the kinetic results are consistent with the activation of eth-
ane being a rate-determining step in the overall oxidation of
ethane. These preliminary kinetic results are consistent with the
previous reports [3,35,38].
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jcat.2011.09.029.
References
[1] D.I. Enache, E. Bordes, A. Ensuque, F. Bozon-Verduraz, Appl. Catal. A 278 (2004)
103.
[
[
2] Y. Wang, K. Otsuka, J. Catal. 171 (1997) 106.
3] Y. Lou, H. Wang, Q. Zhang, Y. Wang, J. Catal. 247 (2007) 245.
[4] Z. Zhao, Y. Yamada, Y. Teng, A. Ueda, K. Nakagawa, T. Kobayashi, J. Catal. 190
2000) 215.
(
It is noted that the selectivity to acetaldehyde from ethane is
lower than that from ethylene, as shown in Fig. 13B. This finding
is in agreement with the previous reports about propane/propyl-
ene oxidation to acrylic acid [39]. It is known that the catalyst sur-
face is more reduced in the oxidation of olefins than in that of
alkanes, which must be linked to different selectivities [40,41].
The higher selectivity to acetaldehyde from the oxidation of ethyl-
ene than from that of ethane may be related to the difference in the
existing states of Mo oxide species. The more easily reduced the
Mo oxide species on the catalyst surface, the higher the selectivity
to acetaldehyde.
[
[
5] F. Wang, J. Dubois, W. Ueda, J. Catal. 268 (2009) 260.
6] A. Erdohelyi, F. Solymosi, J. Catal. 123 (1990) 31.
[7] Z. Zhao, Y. Yusuke, A. Ueda, H. Sakurai, T. Kobayashi, Appl. Catal. A 196 (2000)
7.
[
[
[10] L.J. Huerta, P. Amorós, D. Beltrán-Porter, V.C. Corberán, Catal. Today 117 (2006)
80.
11] Z. Zhao, Y. Yamada, A. Ueda, H. Sakurai, T. Kobayashi, Catal. Today 93–95
2004) 163.
[12] Z. Zhao, J. Liu, C. Xu, A. Duan, T. Kobayashi, I.E. Wachs, Topics Catal. 38 (2006)
09.
3
8] D. Vitry, J. Dubois, W. Ueda, J. Mol. Catal. A 220 (2004) 67.
9] N. Haddad, E. Bordes-Richard, A. Barama, Catal. Today 142 (2009) 215.
1
[
(
3
[
13] D. Zhao, J. Feng, Q. Hu, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Science 279 (1998) 548.
[
[
[
[
14] J.Y. Ying, C.P. Mehnert, M.S. Wong, Angew. Chem. Int. Ed. 38 (1999) 56.
15] Q. Zhang, Y. Wang, Y. Ohishi, T. Shishido, K. Takehira, J. Catal. 202 (2001) 308.
16] J. Liu, Z. Zhao, C. Xu, A. Duan, G. Jiang, J. Phys. Chem. C 112 (2008) 5930.
17] T. Kobayashi, Catal. Today 71 (2007) 69.
4
. Conclusions
In this work, framework-incorporated Mo-SBA-15 mesoporous
[18] R.B. Watson, U.S. Ozkan, J. Catal. 191 (2000) 12.
[
[
[
19] K. Chen, S. Xie, A.T. Bell, E. Iglesia, J. Catal. 198 (2001) 2322.
20] I.E. Wachs, Catal. Today 27 (1996) 437.
21] H. Huang, C. Zhao, Y. Ji, R. Nie, P. Zhou, H. Zhang, J. Hazard. Mater. 178 (2010)
680.
molecular sieve catalysts with different Mo:Si molar ratios were
successfully synthesized by a one-step hydrothermal method. They
were further modified by potassium by the incipient-wetness
impregnation method. The results show that highly ordered meso-
porous structures of Mo-SBA-15 materials are formed with large
surface area and uniform mesoporous structures. XRD, IR, and
Raman results demonstrate that Mo ions are incorporated into
the framework of SBA-15. For both DHT and IWI catalysts, the
structural regularity of SBA-15-based mesoporous material
decreases with increasing Mo content and is destroyed in the
Mo-SBA-15 catalysts at x = 12. Meanwhile, the addition of potas-
sium causes partial collapse of the mesoporous structure of K/
Mo-SBA-15 (initial K:Mo:Si = 0.5:12:100).
[
[
[
22] I. Eswaramoorthi, A.K. Dalai, Micropor. Mesopor. Mater. 93 (2006) 1.
23] L.P. Wang, A.G. Kong, B. Chen, J. Mol. Catal. A 230 (2005) 143.
24] H. Hu, I.E. Wachs, S.R. Bare, J. Phys. Chem. 99 (1995) 10897.
[25] X.T. Gao, S.R. Bare, B.M. Weckhuysen, M. Banares, I.E. Wachs, J. Phys. Chem. B
102 (1998) 10842.
[
[
26] G. Tsilomelekis, A. Christodoulakis, S. Boghosian, Catal. Today 127 (2007) 139.
27] N.F.D. Verbruggen, G. Mestl, L.M.J. Hippel, B. Lengeler, H. Knozinger, Langmuir
10 (1994) 3063.
[28] Z. Liu, Y. Chen, J. Catal. 177 (1998) 314.
[
[
29] N.F.D. Verbruggen, L.M.J. Hippel, G. Mestl, B. Lengeler, H. Knozinger, Langmuir
0 (1994) 3073.
30] R.S. Weber, J. Catal. 151 (1995) 470.
1
[31] A. Christodoulakis, S. Boghosian, J. Catal. 260 (2008) 178.
[
[
32] F. Rahman, K.F. Loughlin, M.A. Al-Saleh, M.R. Saeed, N.M. Tukur, M.M. Hossain,
K. Karim, A. Mamedov, Appl. Catal. A 375 (2010) 17.
33] K. Ruth, R. Burch, R. Kieffer, J. Catal. 175 (1998) 27.
Kinetic data suggest that the selective oxidation of ethane is
approximately first order with respect to ethane concentration
and that ethylene is a possible intermediate for acetaldehyde for-
mation. The activation of ethane may be a rate-determining step
in the overall oxidation of ethane.
[34] M.D. Argyle, K. Chen, A.T. Bell, E. Iglesia, J. Phys. Chem. B 106 (2002) 5421.
[
[
35] E. Heracleous, A.A. Lemonidou, J. Catal. 237 (2006) 175.
36] F. Klose, M. Joshi, C. Hamela, A. Seidel-Morgenstern, Appl. Catal. A 260 (2006)
101.
The potassium-modified catalysts give supercatalytic perfor-
mances for the selective oxidation of ethane to acetaldehyde. TOF
values also increase in the presence of potassium. Among all cata-
lysts, K/Mo-SBA-15 (initial K:Mo:Si = 0.5:10:100) catalyst gives the
[37] M. Machli, C. Boudouris, S. Gaab, J. Find, A.A. Lemonidou, J.A. Lercher, Catal.
Today 112 (2006) 53.
[
[
38] D. Linke, D. Wolf, M. Baerns, S. Zey, U. Dingerdissent, J. Catal. 205 (2002) 32.
39] K. Chen, A. Khodakov, J. Yang, A.T. Bell, E. Iglesia, J. Catal. 186 (1999) 325.
[40] G. Centi, G. Fornasari, F. Trifio, J. Catal. 89 (1984) 44.
[41] B. Solsona, M.I. Vazquez, F. Ivars, A. Dejoz, P. Concepcion, J.M. Lopez Nieto, J.
Catal. 252 (2007) 280.
highest selectivity of CH
3 2 4
CHO + C H (68.3%) and the maximum
yield of acetaldehyde (10.2%). This is the best result for the selec-