Effects of calcination and pretreatment temperatures on the catalytic activity and stability of H2-treated WO3/SiO2 catalysts in metathesis of ethylene and 2-butene

Article abstract of DOI:10.1039/c8ra04949a

The effects of calcination and pretreatment temperatures of the H₂-treated WO₃/SiO₂ catalysts in metathesis of ethylene and 2-butene to propylene were investigated. The results showed that pretreatment with pure hydrogen over the non-calcined catalysts resulted in higher activity and stability than the calcined catalysts, and the hydrogen pretreatment temperature at 650 °C offered the highest 2-butene conversion and propylene selectivity. The calcination of the catalyst before hydrogen pretreatment was proved to be unnecessary. As revealed by various characterization results from N₂ physisorption, XRD, TEM, UV-Vis, Raman, in situ H₂-TPR, in situ NH₃-DRIFTS and in situ NH₃-TPD techniques, activity of the metathesis of ethylene and 2-butene to propylene was related to tungsten dispersion on the support, WO_2.83 and WO₂ phase composition, and isolated surface tetrahedral tungsten oxide species. The stability of the metathesis reaction was also related to the total acidity and the acid sites of both Br?nsted and Lewis acid sites.

Full text of DOI:10.1039/c8ra04949a

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Effects of calcination and pretreatment
temperatures on the catalytic activity and stability
of H -treated WO /SiO catalysts in metathesis of
Cite this: RSC Adv., 2018, 8, 28555
2
3
2
ethylene and 2-butene
Krittidech Gayapan, Sirada Sripinun, Joongjai Panpranot,
and Suttichai Assabumrungrat
Piyasan Praserthdam
*
2 3 2
The effects of calcination and pretreatment temperatures of the H -treated WO /SiO catalysts in metathesis
of ethylene and 2-butene to propylene were investigated. The results showed that pretreatment with pure
hydrogen over the non-calcined catalysts resulted in higher activity and stability than the calcined catalysts,
ꢀ
and the hydrogen pretreatment temperature at 650 C offered the highest 2-butene conversion and
propylene selectivity. The calcination of the catalyst before hydrogen pretreatment was proved to be
unnecessary. As revealed by various characterization results from N
Raman, in situ H -TPR, in situ NH -DRIFTS and in situ NH -TPD techniques, activity of the metathesis of
ethylene and 2-butene to propylene was related to tungsten dispersion on the support, WO2.83 and WO
2
physisorption, XRD, TEM, UV-Vis,
Received 10th June 2018
Accepted 5th August 2018
2
3
3
2
DOI: 10.1039/c8ra04949a
phase composition, and isolated surface tetrahedral tungsten oxide species. The stability of the metathesis
reaction was also related to the total acidity and the acid sites of both Brønsted and Lewis acid sites.
rsc.li/rsc-advances
preparation conditions, support properties, and pretreatment
1
. Introduction
8,10–15
conditions.
Propylene is an important chemical intermediate serving as dispersion, tetrahedral tungsten oxide species, intermediate
16–19
Among them, it was reported that the tungsten
5
+
a building block for petrochemical and plastic products. The WO3ꢁx, or W species are the active sites for metathesis,
demand of propylene has been growing in the global chemical and that WO crystals are not active in metathesis and catalyst
3
1
,2
8,20
industry in recent years. With the increasing demand of sites should be contained in the amorphous surface.
At
propylene, several propylene production technologies including present, the activation by pretreatment conditions to improve
3
4
x 2
catalytic and thermal crackers, methanol to olens, propane performance of supported WO /SiO catalysts in metathesis
5
6
dehydrogenation and olen metathesis, are available. Nowadays, reaction has received much attention because of pre-activating
olen metathesis has become more interesting because it could the active sites of catalysts to form the activated state before
regulate the stock of light olens (ethylene, propylene, and butene) running the reaction.
upon the market demand at low energy and environmental cost.
The olen conversion technology (OCT) process licensed by the propylene self-metathesis activity.
Lummus Technology is the only commercial olen metathesis that H
technology producing propylene for the petrochemical industry, SiO
which used WO /SiO
as a commercial catalyst to convert ethylene WO
2
1,22
Pretreatment of supported WO
3
/SiO
2
7,8
ꢀ
catalysts with CO
2
, H
2
or N
2
gas at around 600 C could improve
23–25
13
Liu et al. reported
ꢀ
content for pretreatment at 400 C on the calcined WO
/
2
3
catalysts caused the catalyst transformed to the tetragonal
3
and partially reduced W2.92, which were active phases for
2
3
2
and 2-butene to propylene in a xed-bed reactor at a temperature > the metathesis of 1-butene and ethylene. In our previous
ꢀ
60 C and pressure of 30–35 bar. The WO /SiO catalysts are the research, we have studied on the effect of pretreatment atmo-
3 2
2
8
2
most widely used catalysts because of their better resistance to sphere over the calcined and non-calcined WO
poisoning, lower price, better stability and easy regeneration.
3
/SiO
for the metathesis of ethylene and 2-butene to propylene by
catalysts for varying the gas pretreatment including pure N , pure H and
, and using the pretreatment temperature of
pretreatment over the non-
catalysts exhibited the WO2.83 phase,
2
catalysts
8,9
The catalytic performance of WO /SiO
3
2
2
2
propylene metathesis is dependent on many parameters mixed H
including the tungsten loading content, tungsten oxide species, 500 C. It was found that pure H
/N
2
2
ꢀ
2
3 2
calcined WO /SiO
which offered higher catalytic performance in the metathesis of
ethylene and 2-butene. However, this research found the
Center of Excellence on Catalysis and Catalytic Reaction Engineering, Department of
Chemical Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok slightly inclined activity through time on stream of 12 h. To
1
0330, Thailand. E-mail: suttichai.a@chula.ac.th; Fax: +66-2218-6877; Tel: +66-
improve the catalytic performance, the research for hydrogen
2
218-6868
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28555

View Article Online
RSC Advances
Paper
pretreatment at elevated temperature have thus been investi- further pretreatment and reaction testing. The calcined catalysts
gated in the present research.
considered by using two calcination temperatures (550 and 650
C) were performed at a desired temperature (550 or 650 C) in air
ꢀ
ꢀ
In this research, the effects of calcination and pretreatment
temperatures on catalytic activities and stability of the H -treated for 8 h, and then the catalysts were packed in a xed bed reactor
2
3 2
WO /SiO catalysts were studied for metathesis of ethylene and 2- for further pretreatment and reaction. For the non-calcined
butene to propylene. The catalyst activity was performed at the catalysts, the samples aer drying were packed in a xed bed
ꢀ
ꢁ1
pressure of 20 bar, temperature of 350 C and WHSV of 0.52 h
The properties of the catalysts were characterized by the Bru-
nauer–Emmett–Teller (BET) method of N physisorption, X-ray
.
reactor for further hydrogen pretreatment and reaction.
2
2.2 Hydrogen pretreatment and catalytic performance
diffraction (XRD), transmission electron microscopy (TEM),
diffuse reectance ultraviolet-visible spectra (UV-Vis DRS),
evaluation
Raman microscopy, diffuse reectance infrared Fourier trans- The metathesis of ethylene and 2-butene to propylene was
form spectroscopy of ammonia with in situ pretreatment (in situ performed in a tubular xed-bed reactor (stainless steel,
NH
with in situ pretreatment (in situ NH
programmed reduction of hydrogen with in situ pretreatment the center of the tubular xed-bed reactor. The sample was
in situ H -TPR) to investigate the relationship of activity and heated to a pretreatment temperature (400, 500, 550, 600, 650 or
stability, textural properties and dispersion of the catalysts, the 700 C) by a furnace under N gas at a ow rate of 30 ml min
3
-DRIFTS), temperature-programmed desorption of ammonia internal diameter ¼ 19.05 mm). The 9% wt WO
3 2
/SiO catalyst
3
-TPD) and temperature- with the calcined or non-calcined catalyst of 3 g was packed at
(
2
ꢀ
ꢁ1
2
structure of surface tungsten compounds, the acidity of the and held at the pretreatment temperature for 1 h. Aer that,
ꢁ
1
catalyst, and the interaction of tungsten species on support.
pure H gas at a total ow rate of 30 ml min was fed to the
2
reactor for 1 h. The reactor was then cooled down to operating
ꢀ
temperature of 350 C under N
2
gas at the same ow rate. The
2
. Experimental
reaction was started by introducing a feed containing 4% 2-
butene (2% cis-2-butene and 2% trans-2-butene) and 8%
2.1 Catalyst preparation
Silica gel, Davisil Grade 646 (40–60 mesh, supplied by “Aldrich”) ethylene balanced in nitrogen gas. The reaction condition was
ꢀ
was used as support. The precursor of 9 wt% WO /SiO catalysts kept at the temperature of 350 C, pressure of 20 bar and WHSV
3
2
ꢁ
1
was prepared by wetness impregnation using an aqueous solu- of 0.52 h . The catalysts were denoted as W-calxxx-Tyyy and W-
tion of ammonium metatungstate hydrate (“Aldrich”, 99.9%). noncal-Txxx representing the H -treated calcined and non-
2
The impregnated sample was dried at room temperature for 2 h calcined catalyst where “xxx” represents the calcination
ꢀ
and subsequently in an oven at 110 C for 24 h. Two types of the temperature including 550 and 650 for the temperature of 550
ꢀ
catalysts as the calcined and non-calcined catalysts were used for and 650 C, while “yyy” represents the pretreatment
Fig. 1 Effect of pretreatment temperature on reaction performance of the calcined and non-calcined WO
3
/SiO
2
catalysts: (a) 2-butene
ꢀ
conversion of the calcined catalysts at 550 C, (b) 2-butene conversion of the non-calcined catalysts, (c) propylene selectivity of the calcined
ꢀ
ꢀ
ꢁ1
catalysts at 550 C, and (d) propylene selectivity of the non-calcined catalysts (T ¼ 350 C, P ¼ 20 bar, WHSV ¼ 0.52 h ).
28556 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
temperature including 400, 500, 550, 600, 650 and 700 for the where C2-butene and S
n
are the 2-butene conversion and product
ꢀ
temperature of 400, 500, 550, 600, 650 and 700 C, respectively. selectivity of n (1 ¼ propylene, 2 ¼ 1-butene, 3 ¼ isobutene, 4 ¼
All of the products were analyzed online by a gas chroma- pentene and 5 ¼ hexene), and where [C
] , [C ]
2-butene out
2
-butene in
tography (Agilent 7820A) with a HP-plot/Al O KCl column, and [C ] are the molar percentage of 2-butene at inlet, 2-butene
2
3
n
which was equipped with a ame ionization detector (FID). The at outlet and product of n (1 ¼ propylene, 2 ¼ 1-butene, 3 ¼
reaction pathways involved for metathesis reaction of ethylene isobutene, 4 ¼ pentene and 5 ¼ hexene), respectively.
and 2-butene to propylene were illustrated in our previous
8
research. The 2-butene conversion and product selectivity were
determined by the following equations:
2.3 Catalyst characterization
½
C
2-buteneꢂ ꢁ ½C2-buteneꢂout
Brunauer–Emmett–Teller (BET) surface areas, pore volumes
and pore sizes of all catalysts were measured by using Micro-
metrics Chemisorbs 2750 with nitrogen adsorption studies. The
X-ray diffraction (XRD) patterns of the catalysts were recorded
in
C
2-butene
¼
½
C
2-buteneꢂin
½
C
n
ꢂ
on a D8 Advance of Bruker AXS using Ni-lter Cu K radiation in
a
S
n
¼
5
ꢀ
ꢀ
P
the 2q range of 20 to 80 . For phase composition identication
purposes, the diffraction patterns were matched with standard
diffraction data in JCPDS les. The Raman spectra of the
½
C
n
ꢂ
n¼1
Fig. 2 Effect of pretreatment temperature on side product selectivity of the calcined and non-calcined WO
3
/SiO
2
catalysts: (a) 1-butene
ꢀ
selectivity of the calcined catalysts at 550 C, (b) 1-butene selectivity of the non-calcined catalysts, (c) isobutene selectivity of the calcined
ꢀ
+
ꢀ
+
5
catalyst at 550 C, (d) isobutene selectivity of the non-calcined catalysts, (e) C
5
] selectivity of the calcined catalysts at 550 C, and (f) C
]
ꢀ
ꢁ1
selectivity of the non-calcined catalysts (T ¼ 350 C, P ¼ 20 bar, WHSV ¼ 0.52 h ).
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28557

View Article Online
RSC Advances
Paper
ꢀ
samples were recorded under ambient conditions using a Sen- sample was cooled down to 40 C under a N
2
ow rate of 25
ꢁ1
terra Dispersive Raman Microscopy (Bruker Optics) equipped ml min . Subsequently, it was exposed to a 15% NH /He mixed
3
ꢁ
1
with the laser wavelength at 532 nm and a TE-cooled CCD gas with a ow rate of 25 ml min for 30 min, and the sample
ꢁ
1
detector. The diffuse reectance ultraviolet-visible spectra (UV- was then purged with a He gas stream (25 ml min ) for 1 h and
Vis DRS) was recorded on Lambda 650 spectrophotometer in the temperature was increased linearly with a rate of
ꢀ ꢀ
ꢁ1
the range between 200 and 800 nm.
10 C min to 500 C. The desorbed ammonia was detected by
The transmission electron microscopy (TEM) images was using the thermal conductivity detector (TCD).
conducted on a JEOL JEM-2010 microscope equipped with
6
a LaB electron gun in the voltage range of 200 kV. The samples
3
. Results and discussion
were prepared by dispersing in ethanol by sonication and a few
drops of suspension onto a carbon-coated copper grid followed
by solvent evaporation in air at room temperature.
3.1 Catalytic performance in the metathesis of ethylene and
2-butenes
The in situ diffuse reectance infrared Fourier transform
spectroscopy (in situ DRIFTS) of ammonia (NH ) adsorption
3
The metathesis performance was illustrated by varying the
pretreatment temperature (400, 500, 550, 600, 650 and 700 C)
ꢀ
spectra at various pretreatment temperature (400, 500, 550, 600,
2
with pure H gas on the calcined and non-calcined catalysts in
ꢀ
6
50 and 700 C) with H
2
gas over the two type catalysts as the
the metathesis of 2-butene (trans- and cis-) and ethylene to
ꢀ
calcined and non-calcined catalysts were recorded with a Bruker
Vertex-70 FT-IR spectrometer equipped with a Harrick Praying
Mantis attachment for diffuse reectance spectroscopy. Diffuse
reectance measurements were performed in situ in a high
temperature cell equipped with a KBr windows. The sample
about 40–50 mg was placed in the Harrick cell, which were cooled
propylene. The reaction was operated at 350 C, 20 bar and 0.52
ꢁ1
h
of WHSV with the reactant of 2% cis-2-butene, 2% trans-2-
butene and 8% ethylene in balanced N . The 2-butene conver-
2
sion and propylene selectivity for the H -treated calcined cata-
2
ꢀ
lysts at a calcination temperature of 550 C and non-calcined
catalysts are shown in Fig. 1. The results showed that
increasing hydrogen pretreatment temperature exhibited
higher 2-butene conversion and propylene selectivity for both
calcined and non-calcined catalysts. In addition, the catalytic
by owing water. The catalyst was heated to a desired pretreat-
ꢀ
ꢀ
ment temperature (400, 500, 550, 600, 650 and 700 C) under a N

2
ꢁ
1
ꢁ1
ow rate of 10 ml min with a heating rate of 10 C min , and
ꢁ1
ꢀ
then the pure H gas at a ow rate of 1 ml min was fed to the
cell for 0.5 h. Aer that, the sample was cooled down to 40 C
under a N ow rate of 10 ml min . Subsequently, it was exposed
/He mixed gas with a ow rate of 10 ml min for
gas stream (10
ml min ) for 1 h. The spectra were collected using a Mercury–
Cadmium–Telluride (MCT) detector with a resolution of 4 cm
and an accumulation of 128 scans.
Temperature-programmed reduction of hydrogen with in
2
performance at the pretreatment temperature of 600 C were
ꢀ
stable for 10 h operation for both calcined and non-calcined
catalysts. In comparison between the calcined catalysts with
ꢁ
1
2
ꢁ
1
ꢀ
to a 15% NH
3
calcination temperature of 550 C and non-calcined catalysts, it
3
0 min, and the sample was then purged with a N
2
2
was found that the H pretreatment of the non-calcined catalyst
ꢁ1
exhibited the 2-butene conversion and propylene selectivity
higher than the calcined catalysts at any pretreatment temper-
ature. The initial 2-butene conversion and propylene selectivity
showed the same trend as the following order: W-noncal-T600 >
W-noncal-T550 > W-cal550-T600 > W-noncal-T500 > W-cal550-
ꢁ
1
situ pretreatment (in situ H -TPR) was performed by using
2
a Micromeritic Chemisorb 2750 automated system with the in
situ pretreatment measurement. The catalyst of 0.1 g was loaded
into a quartz U-tube reactor. Prior to H -TPR experiment, the
2
ꢁ
1
sample was pretreated under a N
a heating rate of 10 C min
2
ꢁ1
ow rate of 25 ml min with
ꢀ
to a desired pretreatment
temperature (400, 500, 550, 600, 650 and 700 C), and then the
pure H gas at a ow rate of 1 ml min was fed to the reactor for
h. Aer that, the reactor was cooled down to 40 C under a N
ꢀ
ꢁ
1
2
ꢀ
1

2
ꢁ1
ow rate of 25 ml min . Subsequently, the catalyst was reduced
ꢁ1
by using 10% H
2
/Ar with a ow rate of 25 ml min at a heating
ꢀ
ꢁ1
ꢀ
rate of 10 C min from 40 to 900 C. The amount of hydrogen
uptake was determined by measuring the areas of the reduction
proles on the thermal conductivity detector (TCD).
Temperature-programmed desorption of ammonia with in
situ pretreatment (in situ NH -TPD) was used to determine the
3
acidity of catalysts by a Micromeritic Chemisorb 2750 auto-
mated system with the in situ pretreatment measurement. The
catalyst of 0.1 g packed in a quartz U-tube reactor was pretreated
ꢁ
1
2
under a N ow rate of 25 ml min with a heating rate of
ꢀ
ꢁ1
1
5
0 C min to a desired pretreatment temperature (400, 500,
Fig. 3 Effect of calcination temperature on reaction performance of
ꢀ
50, 600, 650 and 700 C), and then the pure H
of 1 ml min was fed to the reactor for 1 h. Aer that, the 350 C, P ¼ 20 bar, WHSV ¼ 0.52 h ).
2
gas at a ow rate
catalysts with hydrogen pretreatment at 600 ꢀC (T ¼
the WO
3
/SiO
2
ꢁ1
ꢀ
ꢁ1
28558 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
ꢀ
ꢀ
T550 > W-cal550-T500 > W-noncal-T400 > W-cal550-T400. temperature of 650 C, which was higher than pretreatment
Considering the side product selectivity (Fig. 2), it was found temperature of 600 C, was also investigated. The catalytic
that increasing H pretreatment temperature lowered 1-butene, performances of the calcined catalysts at calcination tempera-
2
+
ꢀ
isobutene, and C ] selectivities but those of the non-calcined tures of 550 and 650 C, and the non-calcined catalysts with
5
ꢀ
catalysts was lower than the calcined catalysts, except the non- hydrogen pretreatment temperature at 600 C are shown in
calcined catalyst with hydrogen pretreatment temperature of Fig. 3. It was found that the calcined catalyst with the calcina-
ꢀ
ꢀ
4
00 C, which exhibited the highest side products.
tion temperature of 550 C exhibited the activity and stability
ꢀ
Interestingly, when considering the results of the hydrogen higher than that of 650 C, but still lower than the non-calcined
pretreatment at 600 C, it was found that the catalytic perfor- catalyst with the same pretreatment temperature. These results
mance and stability of the calcined catalyst (550 C) were close conrmed that the pretreatment temperature of 600 C on the
to the non-calcined catalyst. It should be noted that generally, non-calcined catalyst exhibited the activity and stability of
calcination of the catalysts was used to reduce the effect of metathesis reaction higher than the calcined catalyst, and the
impurities or remove volatile substances from preparation hydrogen pretreatment at a temperature higher than the calci-
ꢀ
ꢀ
ꢀ
26,27
methods,
while catalyst pretreatment was used to obtain the nation temperature also exhibited better activity and stability
22
active sites forming before the reaction running, and the than the higher calcination temperature of the calcined catalyst
pretreatment temperature was usually lower than the calcina- at the same hydrogen pretreatment temperature, and almost
tion temperature. In further studies, the calcination close to the non-calcined catalyst. Therefore, the improvement
of metathesis reaction of ethylene and 2-butene could be ach-
ieved by hydrogen pretreatment of the non-calcined catalysts or
calcined catalyst with a calcination temperature lower than the
pretreatment temperature. Fig. 4 shows the experimental
results of the non-calcined catalysts at higher pretreatment
ꢀ
temperatures (650 and 700 C). It was found that the optimum
ꢀ
hydrogen pretreatment temperature was at 650 C, in which the
highest 2-butene conversion and propylene selectivity, and the
lowest side product selectivities including 1-butene, isobutene
+
and C ] were obtained. Such results suggested that the
5
elevated temperature for hydrogen pretreatment over the
calcined and non-calcined catalysts obviously affected the
activity and stability of the catalysts, and the non-calcined
ꢀ
catalyst pretreated with hydrogen at the temperature of 650 C
exhibited the highest activity and stability on the metathesis of
ethylene and 2-butene to propylene.
3
.2 Textural properties and dispersion of the catalysts
The N adsorption–desorption isotherms and pore size distri-
bution of the H pretreatment on the calcined and non-calcined
catalysts at different pretreatment temperatures are shown in
2
Fig.
4
Catalytic performance of hydrogen pretreatment under
ꢀ
different temperatures of 600, 650 and 700 C of the non-calcined
WO
2
ꢀ
ꢁ1
3
/SiO
2
catalysts (T ¼ 350 C, P ¼ 20 bar; WHSV ¼ 0.52 h ).
Fig. 5
dashed lines) WO
d) W-noncal-T500, (e) W-cal550-T550, (f) W-noncal-T550, (g) W-cal550-T600, (h) W-noncal-T600, (i) W-cal650-T600, (j) W-noncal-T650
and (k) W-noncal-T700.
N
2
adsorption–desorption isotherms (left) and pore size distributions (right) of the calcined (continuous lines) and non-calcined (short
3
/SiO catalysts pretreated with pure H at different temperatures; (a) W-cal550-T400, (b) W-noncal-T400, (c) W-cal550-T500,
2
2
(
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28559

View Article Online
RSC Advances
Paper
Fig. 5, and their textural properties are listed in Table 1. The temperatures showed the XRD patterns differently based on the
isotherms of all catalysts showed typical IV isotherms with the standard reference data in JCPDS. For pretreatment tempera-
ꢀ
H1 hysteresis loop, which revealed that all catalysts possessed ture at 400 C, the XRD patterns were assigned to tetragonal
2
8,29
a mesoporous structure.
The sharp capillary condensation WO for the calcined catalysts and amorphous WO for the non-
3 3
ꢀ
feature at relative pressure of 0.75–0.95 (P/P
0
) indicated the calcined catalysts, while those pretreated at 500 and 550 C
existence of uniform pores in all catalysts. It could be clearly showed the WO2.92 phase for the calcined catalyst and WO2.83
observed that all catalysts showed similar narrow pore size phase for the non-calcined catalyst, and those pretreated at
ꢀ
distributions (Fig. 5(right)). The specic surface area, pore equal or higher than 600 C showed the multiple phase
volume, and average pore size of all catalysts (in Table 1) compositions for both calcined and non-calcined catalysts. The
ꢀ
showed no signicant differences within experimental error hydrogen pretreatment at 600 C for the calcined catalysts with
ꢀ
accepted around 5–10%. In other words, the pretreatment the calcination temperature of 550 and 650 C showed the
temperature did not affect much the physical properties of the WO2.92, WO2.83 and WO2 phase composition, whereas the
ꢀ
calcined and non-calcined catalysts.
patterns of the calcination at 550 C exhibited lower crystal-
The XRD patterns of all pretreated catalysts are shown in linity, indicating the better dispersed tungsten species on the
8
Fig. 6. The catalysts obtained at various hydrogen pretreatment surface. For the hydrogen pretreatment at equal and higher
ꢀ
than 600 C on the non-calcined catalysts, the XRD patterns for
ꢀ
the pretreatment temperature of 600 C showed the WO2.83 and
Table 1 BET surface areas, pore volumes and pore sizes of different
WO /SiO catalysts
ꢀ
3
2
WO phase composition and that of 650 C showed the WO2.83,
WO and W phase composition, while that of 700 C showed
the WO and W phase composition. In addition, it was found
that the XRD patterns of the non-calcined catalyst at pretreat-
ment of 650 C exhibited lower crystallinity than those of other
pretreatment temperatures, indicating the higher dispersion
2
0
ꢀ
2
BET surface
area (m g
Pore volume
Pore size
(nm)
0
2
ꢁ1
3
ꢁ1
2
Catalysts
)
(cm g
)
ꢀ
W-cal550-T400
W-cal550-T500
W-cal550-T550
W-cal550-T600
W-cal650-T600
W-noncal-T400
W-noncal-T500
W-noncal-T550
W-noncal-T600
W-noncal-T650
W-noncal-T700
286
287
289
294
266
288
298
300
316
303
296
0.98
1.04
0.98
1.05
0.98
0.97
1.03
1.01
1.07
1.02
1.02
9.35
9.21
9.19
8.85
9.53
9.49
9.45
8.98
9.08
9.15
9.48
26
of W species on the support.
The particle morphology and lattice spacing of the samples
were investigated by TEM and HRTEM images and the results
are shown in Fig. 7. For the TEM images, it was found that
increasing hydrogen pretreatment temperature exhibited the
better tungsten dispersion on the surface catalysts, as indicated
by fewer agglomerates than the lower pretreatment tempera-
ture. For any pretreatment temperature, the H
2
-treated non-
calcined catalysts showed fewer agglomerates than the H
2
-
Fig. 6 XRD patterns of the calcined (colored continuous lines) and non-calcined (short dashed lines) WO
pretreatment at different temperatures and standard reference data from JCPDS (black continuous lines).
3 2 2
/SiO catalysts with pure H
28560 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
treated calcined catalysts. As revealed by the high resolution the spectra for all the catalysts (inset gure), whereas the adsorp-
TEM images (inset gure), it was found only one kind of the tion bands between 400 and 800 nm were also observed for all the
8
,15,17
crystal lattice planes for the H -treated calcined and non- catalysts. According to the literatures,
the absorption bands at
2
ꢀ
2ꢁ
calcined catalysts at temperature of 400, 500 and 550 C, 230 nm could be assigned to isolated tetrahedral [WO4] species,
except the H
2
-treated non-calcined catalyst at temperature of while the absorption bands at 270 nm corresponded to octahedral
ꢀ
400 C, in which the crystal lattice planes were not found due to polytungstate species. The adsorption bands between 400 and
8
4+
5+
17
2
its amorphous structure. Whilst for the H -treated calcined and 800 nm could be assigned to W and W species. As shown in
non-calcined catalysts at temperature of 600 C, various crystal this gure, it was found that increasing the hydrogen pretreatment
lattice planes were found. Nevertheless, the pretreatment temperature affected the intensity of adsorption band at both 230
ꢀ
ꢀ
temperature at 650 and 700 C of the non-calcined catalyst and 270 nm slightly increased, and these adsorption bands of the
illustrated the three and two kinds of the crystal lattice planes H -treated non-calcined catalysts exhibited the intensities higher
2
also. The lattice spacing of all catalysts revealed the spacing at than the H -treated calcined catalysts. The adsorption band
2
around 0.22, 0.34, 0.38 and 0.39 nm, which were consistent with between 400 and 800 nm of the catalysts was also broader with
2
the XRD standard reference data in JCPDS. The lattice spacing increasing the pretreatment temperature, and that of the H -
ꢀ
of the (001) planes of tetragonal WO
3
, (010) planes of WO2.92
,
treated non-calcined catalysts at 700 C showed the broadest ones,
(
010) planes of WO2.83, (011) planes of WO
2
and (110) planes of ascribed to the ordered mesoporous structure and enlarged oxygen
0
5+
30
W were about 0.392, 0.382, 0.379 and 0.343 nm, respectively.
vacancies due to higher W in sub-stoichiometric WO3ꢁx, which
were consistent with the XRD results.
The Raman spectra of the H -treated calcined and non-calcined
2
3
.3 Structure of surface tungsten compounds
catalysts at different pretreatment temperatures was also shown in
ꢁ
1
Structure of tungsten species produced by pure H
2
gas pretreat- Fig. 9. The bands at 263–275, 325–329, 701–720 and 801–810 cm ,
ment at different pretreatment temperatures on the calcined and which were identied as the four strongest modes of tungsten
30
non-calcined catalysts were analyzed by UV-Vis DRS as shown in oxide, were assigned to the deformation mode of W–O–W,
Fig. 8. Two absorption bands at 230 and 270 nm were observed in bending mode of O–W–O, bending mode of W–O, and symmetric
3 2 2
Fig. 7 TEM and HRTEM (inset) images of the calcined (top) and non-calcined (bottom) WO /SiO catalysts with pure H pretreatment at different
temperatures; (a) W-cal550-T400, (b) W-noncal-T400, (c) W-cal550-T500, (d) W-noncal-T500, (e) W-cal550-T550, (f) W-noncal-T550, (g) W-
cal550-T600, (h) W-noncal-T600, (i) W-cal650-T600, (j) W-noncal-T650 and (k) W-noncal-T700.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28561

View Article Online
RSC Advances
Paper
pretreatment temperature due to an increasing amount of oxygen
30
vacancies in the form of WO3 phase, which were consistent
ꢁx
with the XRD and UV-Vis results. In addition, these bands of the
non-calcined catalysts were broader than the calcined catalysts,
indicating that higher oxygen vacancies occurring on the WO
3
8
ꢁ1
crystalline. Considering the Raman band at 970 cm of the
calcined and non-calcined catalysts, it was found that this band of
the non-calcined catalyst showed more obvious peaks than that of
the calcined samples, indicating more isolated surface tetrahedral
tungsten oxide species as also conrmed by the UV-Vis results. In
addition, higher calcination temperature caused low intensity of
ꢁ
1
the Raman band at 970 cm . The comparison of the ratios of
ꢁ
1
relative Raman intensities of the peak at 970 to 805 cm (I970/I805
)
are shown in table (inset in Fig. 9), which was an indicative of the
relative amount of isolated surface tetrahedral tungsten oxide
8,33
species (active sites) to WO
3
crystal (non-active sites). The high
Fig. 8 UV-Vis DRS patterns of the calcined (continuous lines) and
pretreatment temperature exhibited the higher I970/I805 ratio in
3 2
non-calcined (short dashed lines) WO /SiO catalysts pretreated with
pure H
2
at different temperatures; (a) W-cal550-T400, (b) W-noncal- both calcined and non-calcined catalysts and the H pretreatment
2
ꢀ
T400, (c) W-cal550-T500, (d) W-noncal-T500, (e) W-cal550-T550, (f)
W-noncal-T550, (g) W-cal550-T600, (h) W-noncal-T600, (i) W-
cal650-T600, (j) W-noncal-T650 and (k) W-noncal-T700.
at 650 C of the non-calcined catalyst exhibited the highest I970/I805
ratio among the catalysts studied. Based on the characterization
results described above, it was concluded that H pretreatment of
2
the non-calcined catalysts exhibited higher amount of isolated
1
5,30,31
surface tetrahedral tungsten oxide than the calcined ones and the
stretching mode of W–O, respectively.
The broad band at
ꢁ1
ꢀ
H
2
pretreatment at 650 C of the non-calcined catalyst showed the
9
70 cm was assigned to the O]W]O band of isolated surface
tetrahedral tungsten oxide species. The hydrogen pretreatment at
32
highest amount of isolated surface tetrahedral tungsten oxide.
different temperatures of all catalysts showed the four strongest
ꢁ
1
peaks (263–275, 325–329, 701–720 and 801–810 cm ) of crystal-
3
.4 Acidity of the catalysts
To distinguish the differences in NH
the catalysts, induced by the hydrogen pretreatment tempera-
ture of the calcined and non-calcined WO /SiO catalysts, the in
ꢁ
1
line WO
3
but the small broad peaks at 970 cm were different.
became broader with
3
adsorption behaviors over
The four strongest peaks of crystalline WO
3
increasing the pretreatment temperature of the H -treated
2
calcined and non-calcined catalysts, suggesting that there was
a gradual degradation of the crystallinity upon increasing
3
2
situ DRIFTS spectra of ammonia adsorption/desorption were
performed under different pretreatment temperatures. Fig. 10
shows the NH adsorption in the region of 1100–1800 cm over
3
ꢁ
1
the H
2
-treated calcined and non-calcined catalysts at NH
3
ꢀ
desorption temperature of 40 C. Several kinds of NH
3
species
corresponding to different adsorption wavenumbers were
observed aer the catalysts exposed to ammonia gas. The bands
ꢁ1
at 1463 and 1685 cm could be assigned to Brønsted acid sites,
ꢁ
1
while the bands at 1230, 1280, 1315 and 1615 cm could be
34–36
assigned to Lewis acid sites,
and the acid amount of the total
Brønsted (TB) and Lewis (TL) acid sites are shown in Fig. 11 and
Table 2. It was found that increasing pretreatment temperature
increased both Brønsted and Lewis acid sites for the calcined
catalysts due to the formation of Si–O–W–OH species through
37
the reaction of terminal silanols and surface WO . However, at
3
ꢀ
pretreatment temperature of 400 C, the catalysts showed
higher Lewis acid sites than the others. On the other hand,
those of the non-calcined catalysts decreased with increasing of
the hydrogen pretreatment temperature, which indicated that
the formation of the Si–O–W–OH species for the non-calcined
3
catalysts was probably due to the higher dispersion of WO on
the support, which was consistent with the BET, XRD, and
Fig. 9 Raman spectra of the calcined (continuous lines) and non-
calcined (short dashed lines) WO
at different temperatures; (a) W-cal550-T400, (b) W-noncal-T400,
c) W-cal550-T500, (d) W-noncal-T500, (e) W-cal550-T550, (f) W-
3 2
/SiO catalysts pretreated with pure
Raman results. Comparing the calcination temperature
H
2
ꢀ
between 550 and 650 C for the H -treated calcined catalysts, it
2
(
was found that higher calcination temperature caused low
Brønsted and Lewis acid sites. Considering the ratios of
noncal-T550, (g) W-cal550-T600, (h) W-noncal-T600, (i) W-cal650-
T600, (j) W-noncal-T650 and (k) W-noncal-T700.
28562 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
2 2
displayed in Fig. 13. The H -TPR proles of the H -treated
calcined and non-calcined WO /SiO catalysts showed three
3
2
ꢀ
peaks at 495–505, 565, and 780–825 C. The low temperature
peaks were attributed to the reduction of W species in octahe-
dral coordination and the high temperature peak was assigned
1
5,17
to the reduction of the surface amorphous WO
Comparing between the calcined and non-calcined WO
3
species.
/SiO
3
2
catalysts, the reduction peak at low temperature of the non-
calcined catalysts appeared at higher temperature than the
calcined ones, indicating that interaction of W species in octa-
hedral coordination on silica support of the non-calcined
catalysts was stronger than the calcined catalysts. Whilst the
reduction peaks at high temperature of the non-calcined cata-
lysts appeared at lower temperature than the calcined catalysts,
indicating that reduction of WO
3
crystal of the non-calcined
catalysts was weaker than the calcined catalysts, which re-
15

ected an improved reducibility of dispersed species. In
addition, comparing the calcination temperature of 550 and
Fig. 10 In situ NH
3
-DRIFTS profiles of the calcined (continuous lines)
/SiO catalysts pretreated
ꢀ
6
50 C of the calcined catalysts, it was found that reduction
and non-calcined (short dashed lines) WO
with pure H
3
2
ꢀ
at different temperatures; (a) W-noncal-T400, (b) W- peak at low temperature of the calcination of 650 C appeared at
noncal-T500, (c) W-noncal-T550, (d) W-noncal-T600, (e) W-noncal- higher temperature than that of 550 C, while the reduction
2
ꢀ
T650, (f) W-noncal-T700, (g) W-cal550-T400, (h) W-cal550-T500, (i)
W-cal550-T550, (j) W-cal550-T600 and (k) W-cal650-T600.
peak at high temperature was not signicantly different. Inter-
estingly, when the pretreatment temperature of the non-
ꢀ
calcined catalysts higher than 600 C, it was found that the
reduction peak at low temperature shied to higher tempera-
Brønsted and Lewis acid sites (Table 2), it was found that the
elevated pretreatment temperature of both H -treated calcined
ꢀ
ture, and the pretreatment of 650 C shied the low tempera-
2
ꢀ
ture peak to higher temperature than that of 700 C, indicating
and non-calcined catalysts was likely to increase these ratios.
15
that higher interaction of tungsten species on the support. In
ꢀ
And comparing the calcination temperature of 550 and 650 C
contrast, reduction peak at high temperature for the pretreat-
ꢀ
for the calcined catalysts pretreated at 600 C, the calcination of
ꢀ
ꢀ
ment temperature of 650 and 700 C was slightly broad, and the
ꢀ
ꢀ
6
50 C exhibited these ratios higher than that of 550 C.
pretreatment temperature of 700 C was broader and shied to
To investigate the total acidity of the catalyst, the NH
3
-TPD
ꢀ
lower temperature than that of 650 C, which might be assigned
was performed over the H
catalysts with different pretreatment temperatures. The NH
TPD proles are displayed in Fig. 12. The ammonia desorption
peaks of the non-calcined WO /SiO catalysts occurred at higher
temperature than the calcined catalysts, indicating the higher
2
-treated calcined and non-calcined
0
15
0
to higher W formation and was conrmed the W occurring
by the XRD results. Considering the effect of hydrogen
pretreatment temperature on the calcined and non-calcined
catalysts, the increasing of the pretreatment temperature up
3
-
3
2
ꢀ
to 600 C led to a boarder low temperature peak, while the high
8,32
acid strength as stronger acidity. In addition, increasing the
pretreatment temperature caused the ammonia desorption
peak to slightly shi to lower temperature, indicating lower acid
temperature peaks of either calcined or non-calcined catalysts
were not signicantly different, and the low H consumption
2
peaks (inset gure) shied to lower temperature, except the
8
strength. Considering the amount of total acidity of all the
ꢀ
pretreatment temperature of 600 C on the calcined catalyst at
catalysts as shown in Table 2, it was found that the amounts of
ꢀ
5
50 C, which exhibited two peaks at low temperature and the
total acidity of the H
at 550 C increased with increasing pretreatment temperature
2
-treated calcined catalysts for calcination
lowest temperature peaks were shied to higher temperature.
ꢀ
ꢀ
In addition, pretreatment at 600 C on the catalyst calcined at
ꢀ
(except for pretreatment temperature at 400 C that showed the
ꢀ
6
50 C could improve the interaction of tungsten species on the
ꢀ
total acidity amount higher than that at 500 and 550 C, but less
support, comparing with the same pretreatment temperature
ꢀ
than that at 600 C), while those of the non-calcined catalysts
ꢀ
on the calcined catalyst at 550 C. Furthermore, increasing
decreased with increasing the pretreatment temperature, which
ꢀ
pretreatment temperature higher than 600 C for the H
2
-treated
were consistent with the results of the in situ DRIFTS. In addi-
0
non-calcined catalyst also led to higher interaction and W
formation, which was consistent with the XRD, UV-Vis, and
Raman results.
ꢀ
tion, comparing the calcination temperature of 550 and 650 C
for the calcined catalysts, it was found that higher calcination
temperature caused low total amount of acidity, which was
consistent with the in situ DRIFTS results.
3.6 Structural–activity relationship
3.5 Interaction of tungsten species on support
The effects of hydrogen pretreatment over the calcined and non-
The H -TPR was performed to investigate the interaction calcined WO
between W active species and silica support and the proles are ethylene and 2-butene at 20 bar and 350 C. Different
2
3
/SiO
2
catalysts were studied in the metathesis of
ꢀ
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28563

View Article Online
RSC Advances
Paper
Fig. 11 The acid amount of Brønsted and Lewis acid sites of the calcined and non-calcined WO
different temperatures.
3 2 2
/SiO catalysts pretreated with pure H at
a
Table 2 The amount of total acidity, Brønsted and Lewis acid sites of different WO
3
/SiO
2
catalysts
b
ꢁ1
c
ꢁ1
c
ꢁ1
Samples
Total acidity (mmol g
)
Total Brønsted acid (mmol g
)
Total Lewis acid (mmol g
)
TB/TL
W-cal550-T400
W-cal550-T500
W-cal550-T550
W-cal550-T600
W-cal650-T600
W-noncal-T400
W-noncal-T500
W-noncal-T550
W-noncal-T600
W-noncal-T650
W-noncal-T700
2.20
2.06
2.13
2.35
1.68
4.49
3.73
3.25
2.72
2.32
1.54
0.84
1.12
1.19
1.29
1.11
1.88
1.86
1.59
1.53
1.33
0.87
1.16
0.94
0.96
1.03
0.56
1.67
1.61
1.36
1.15
0.98
0.51
0.72
1.20
1.23
1.26
2.00
1.12
1.16
1.17
1.33
1.36
1.71
a
b
c
TB ¼ total Brønsted acid sites and TL ¼ total Lewis acid sites. From in situ NH
3
-TPD data calculation. From in situ NH
3
-DRIFTS data calculation.
28564 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
hydrogen pretreatment at elevated temperature was not much
signicant to affect the activity for the metathesis of ethylene
and 2-butene. Considering the dispersion of tungsten on the
silica support (as conrmed by XRD and TEM results), it was
2
found that the H -treated non-calcined catalysts exhibited the
activity higher than the H
pretreatment temperature at 650 C of the H
2
-treated calcined catalysts, and the
ꢀ
2
-treated non-
calcined catalyst showed the highest activity, indicating that
better dispersion of tungsten on the silica support exhibited the
high activity of metathesis of ethylene and 2-butene to produce
14,26,38
propylene.
Moreover, the TEM mapping results also
conrmed that the H -treated calcined catalysts showed the
2
more tungsten agglomerates dispersed on the surface silica
support than the H -treated non-calcined catalysts, indicating
2
poor tungsten dispersion on the support. Additionally, the UV-
Vis and Raman results indicated that the higher amount of
isolated tetrahedral tungsten oxide was active species for
8,17
metathesis reaction.
Considering the crystalline phase composition and average
Fig. 12 NH
calcined (short dashed lines) WO
3
-TPD profiles of the calcined (continuous lines) and non-
/SiO catalysts with pure H
3
2
2
pretreatment at different temperatures; (a) W-noncal-T400, (b) W- crystal size of the catalysts calculated by the Scherrer equation
noncal-T500, (c) W-noncal-T550, (d) W-noncal-T600, (e) W-noncal- for the XRD results as shown in Table 3, it was found that
T650, (f) W-noncal-T700, (g) W-cal550-T400, (h) W-cal550-T500, (i)
W-cal550-T550, (j) W-cal550-T600 and (k) W-cal650-T600.
2
increasing the pretreatment temperature of the H -treated
calcined and non-calcined catalyst caused the different multiple
phase compositions and decreased the average crystal size.
When comparison with the initial 2-butene conversion,
propylene and 1-butene selectivity (in Table 3) at operation of
pretreatment temperatures could lead to the changes of the
textural properties and dispersion of the WO /SiO catalysts, the
3 2
1
h, it was found that the composition WO2.83 and WO phase
2
structure of surface tungsten compounds, the acidity of the
catalyst, and the interaction of tungsten species on support. As
revealed on the BET results, the specic surface area, pore
volume and average pore size of all the catalysts occurring from
on the pretreated catalysts caused the high 2-butene conversion
and propylene selectivity, and lower the 1-butene selectivity. In
addition, small crystal size of these phases also exhibited the
high catalytic activity, which was related with higher tungsten
Fig. 13
2 3 2 2
H -TPR patterns of the calcined (continuous lines) and non-calcined (short dashed lines) WO /SiO catalysts with in situ pure H
pretreatment at different temperatures; (a) W-cal550-T400, (b) W-noncal-T400, (c) W-cal550-T500, (d) W-noncal-T500, (e) W-cal550-T550, (f)
W-noncal-T550, (g) W-cal550-T600, (h) W-noncal-T600, (i) W-cal650-T600, (j) W-noncal-T650 and (k) W-noncal-T700.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28565

View Article Online
RSC Advances
Paper
Table 3 The detailed parameters of XRD results and catalytic performance of 2-butene conversion, propylene and 1-butene selectivities at 1 h
over the calcined and non-calcined WO
3
/SiO
2
catalysts pretreated with pure H
2
at different temperatures
Crystal size
(nm)
2-Butene
conversion (%)
Propylenea
1-Butene a
Isobutene
selectivity (%)
ꢀ
a
a
Catalysts
Phase
2q ( )
selectivity (%)
selectivity (%)
W-cal550-T400
W-cal550-T500
W-cal550-T550
W-cal550-T600
Tetragonal WO
WO2.92
WO2.92
WO2.92
WO2.83
3
22.9
23.3
23.2
n.a.
23.5
25.8
n.a.
23.4
25.7
n.a.
23.4
23.4
23.5
25.9
23.4
25.8
40.3
25.6
40.1
19.8
16.3
15.8
n.a.
15.2
33.3
n.a.
15.9
35.1
n.a.
14.5
14.2
13.0
32.4
10.8
31.8
21.2
34.3
25.1
24.4
46.2
57.3
71.0
7.9
75.3
84.6
90.5
80.2
16.9
9.9
8.2
2.2
1.0
0.8
5.6
WO
2
W-cal650-T600
WO2.92
WO2.83
68.4
88.0
7.9
0.8
WO
2
W-noncal-T400
W-noncal-T500
W-noncal-T550
W-noncal-T600
Amorphous WO
WO2.83
WO2.83
3
52.6
68.8
72.7
75.3
59.8
86.2
89.9
91.6
18.6
7.3
4.9
4.4
10.4
1.5
1.2
1.1
WO2.83
WO
2
W-noncal-T650
WO2.83
75.4
68.1
92.4
90.8
4.2
6.5
0.9
0.2
WO
W
2
2
0
W-noncal-T700
WO
0
W
a
Data in operation of 1 h.
14,26
dispersion on the support.
Therefore, the two mentioned (Brønsted and Lewis acid sites) played an important role to the
crystalline phases are good proxy (indicators) of the reducibility stability in the metathesis of ethylene and 2-butene.
of the W oxide phase, which itself is related to dispersion. Considering the interaction of tungsten species on the
Considering the acidity of catalysts (in situ NH -DRIFTS and support (in situ H -TPR results), it was found that the H -treated
2
NH -TPD results), it was found that the suitable amount of total non-calcined catalysts exhibited stronger interaction than the H -
3
2
2
3
acidity and acid sites (Brønsted and Lewis acid sites) exhibited treated calcined catalysts at any pretreatment temperature, and
ꢀ
the high stability on the metathesis reaction of ethylene and 2- increasing the pretreatment temperature to 600 C caused the
butene to propylene, in which pretreatment of the non-calcined interaction between tungsten species and silica support
ꢀ
ꢀ
ꢀ
catalyst at 600 and 650 C, and the calcined catalyst at 550 C decreased for both calcined catalyst at calcination of 550 C and
ꢀ
with pretreatment of 600 C showed the best stability through non-calcined catalyst. However, pretreatment temperature higher
0 h operation. These results were supported by the literatures than 600 C caused the higher interaction between tungsten
ꢀ
1
that reported that the increased Brønsted acid sites could help species and support. In addition, comparing the catalytic
to promote the metathesis reaction of ethylene and 2-butene to performances, it was found that the interaction between tungsten
1
2,17,37,39
propylene,
but high number of the Brønsted acid sites species and silica support occurred from the hydrogen pretreat-
37
led to isomerization of 1-butene to isobutene. On the other ment at elevated temperature on the catalysts did not have
hand, increased Lewis acid sites could help to promote the signicant impact on activity and stability on metathesis reaction
3
7,39
isomerization of 1-butene to 2-butene,
and suitable amount of ethylene and 2-butene to propylene. Nevertheless, considering
ꢀ
of acid sites was important for the stable activity of olen the hydrogen pretreatment at the temperature of 600 C on the
metathesis. In addition, crystalline WO
isomerization reactions, and easily resulted in the deactivation pretreatment temperature higher than the calcination tempera-
39
ꢀ
3
could catalyze some calcined catalyst at calcination of 550 C, it was found that the
39,40
of catalysts.
In the catalytic performance for 1 h operation as ture exhibited the catalytic performances and characterization
shown in Table 3, it was found that the H -treated calcined close to the non-calcined catalyst at the same pretreatment
2
ꢀ
catalyst at pretreatment temperature under 550 C exhibited temperature. Such results indicated that the pretreatment
high isomerization, indicating that high crystalline WO caused temperature on the calcined catalysts at higher than the calci-
3
ꢀ
high isomerization reaction of 2-butene to 1-butene and that of nation temperature (550 C) could improve the catalytic perfor-
1
-butene to isobutene, while the H
2
-treated non-calcined cata- mances to be close to the non-calcined catalyst, which was more
ꢀ
lysts at pretreatment temperature under 550 C also exhibited active in the metathesis of ethylene and 2-butene to produce
high isomerization, indicating that high Lewis and Brønsted propylene. In other words, the catalytic performances in
acid sites caused high isomerization reaction of 2-butene to 1- metathesis of ethylene and 2-butene to produce propylene could
butene and that of 1-butene to isobutene, respectively. There- be improved by using the hydrogen pretreatment at elevated
fore, the suitable amount of total acidity and acid sites temperature on the non-calcined catalyst.
28566 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
1
1 T. I. Bhuiyan, P. Arudra, M. M. Hossain, M. N. Akhtar,
A. M. Aitani, R. H. Abudawoud and S. S. Al-Khattaf, Can. J.
Chem. Eng., 2014, 92, 1271–1282.
4
. Conclusions
The hydrogen pretreatment at elevated temperature of the
calcined and non-calcined WO
reaction of ethylene and 2-butene to propylene was investigated.
The catalytic activity and stability could be improved by the 13 H. Liu, K. Tao, H. Yu, C. Zhou, Z. Ma, D. Mao and S. Zhou, C.
hydrogen pretreatment. The results showed that the non-
R. Chim., 2015, 18, 644–653.
3
/SiO
2
catalysts for metathesis 12 S. Chaemchuen, S. Phatanasri, F. Verpoort, N. Sae-ma and
K. Suriye, Kinet. Catal., 2012, 53, 247–252.
calcined catalysts with hydrogen pretreatment temperature of 14 N. Poovarawan, K. Suriye, J. Panpranot, W. Limsangkass,
ꢀ
6
50 C offered the highest activity and stability. The calcination
F. J. Santos Cadete Aires and P. Praserthdam, Catal. Lett.,
2015, 145, 1868–1875.
of the catalyst was unnecessary and it showed adverse effects
when the calcination temperature was higher than the 15 Q. Zhao, S.-L. Chen, J. Gao and C. Xu, Transition Met. Chem.,
hydrogen pretreatment temperature. The characterization
2009, 34, 621–627.
results indicated that the better tungsten dispersion, WO2.83 16 S. J. Choung and S. W. Weller, Ind. Eng. Chem. Process Des.
and WO phase compositions, and higher isolated surface
Dev., 1983, 22, 662–665.
tetrahedral tungsten oxide species are favorable to the high 17 S. Huang, S. Liu, W. Xin, J. Bai, S. Xie, Q. Wang and L. Xu, J.
activity of the metathesis reaction of ethylene and 2-butene to
Mol. Catal. A: Chem., 2005, 226, 61–68.
propylene, while the suitable amount of total acidity and acid 18 H. Liu, S. Huang, L. Zhang, S. Liu, W. Xin and L. Xu, Catal.
sites of both Brønsted and Lewis acid sites played an important
Commun., 2009, 10, 544–548.
2
role to the stability of metathesis reaction. In addition, the 19 A. J. Van Roosmalen and J. C. Mol, J. Catal., 1982, 78, 17–23.
textural properties and interaction between W species and silica 20 S. Huang, F. Chen, S. Liu, Q. Zhu, X. Zhu, W. Xin, Z. Feng,
support occurring from pretreatment temperature were not
signicant to the activity and stability.
C. Li, Q. Wang and L. Xu, J. Mol. Catal. A: Chem., 2007,
267, 224–233.
21 K. Ding, A. Gulec, A. M. Johnson, T. L. Drake, W. Wu, Y. Lin,
E. Weitz, L. D. Marks and P. C. Stair, ACS Catal., 2016, 6,
Conflicts of interest
5740–5746.
There are no conicts of interest to declare.
22 S. Lwin and I. E. Wachs, ACS Catal., 2017, 7, 573–580.
2
3 A. Andreini and J. C. Mol, J. Colloid Interface Sci., 1981, 84,
7–65.
5
Acknowledgements
24 R. C. Luckner, G. E. McConchie and G. B. Wills, J. Catal.,
1973, 28, 63–68.
The authors acknowledge the support from the Department of
Chemical Engineering, Faculty of Engineering, Chulalongkorn 25 R. Westhoff and J. A. Moulijn, J. Catal., 1977, 46, 414–416.
University. Department of Chemical Engineering, Faculty of 26 S. Maksasithorn, P. Praserthdam, K. Suriye and
Engineering, Chulalongkorn University would like to acknowl-
edge The Institutional Research Grant (The Thailand Research
D. P. Debecker, Microporous Mesoporous Mater., 2015, 213,
125–133.
Fund), IRG 5780014, and Chulalongkorn University, Contract 27 J. Thomas Richardson, Principles of catalyst development,
No. RES_57_411_21_076.
1991.
28 C. Lin, K. Tao, H. Yu, D. Hua and S. Zhou, Catal. Sci. Technol.,
2
014, 4, 4010–4019.
References
2
9 H. Liu, K. Tao, P. Zhang, W. Xu and S. Zhou, New J. Chem.,
2015, 39, 7971–7978.
30 L. Wang, Y. Wang, Y. Cheng, Z. Liu, Q. Guo, M. N. Ha and
1
W. Jiang, R. Huang, P. Li, S. Feng, G. Zhou, C. Yu, H. Zhou,
C. Xu and Q. Xu, Appl. Catal., A, 2016, 517, 227–235.
J. C. Mol, J. Mol. Catal. A: Chem., 2004, 213, 39–45.
2
3
Z. Zhao, J. Mater. Chem. A, 2016, 4, 5314–5322.
J. Towghi, A. Niaei, R. Karimzadeh and G. Saedi, Korean J. 31 Y. Wang, Q. Chen, W. Yang, Z. Xie, W. Xu and D. Huang,
Chem. Eng., 2006, 23, 8–16.
Appl. Catal., A, 2003, 250, 25–37.
4
5
M. St ¨o cker, Microporous Mesoporous Mater., 1999, 29, 3–48.
F. Cavani, N. Ballarini and A. Cericola, Catal. Today, 2007,
32 S. Maksasithorn, D. P. Debecker, P. Praserthdam,
J. Panpranot, K. Suriye and S. K. N. Ayudhya, Chin. J.
Catal., 2014, 35, 232–241.
127, 113–131.
6
7
8
9
X. Li, W. Zhang, S. Liu, L. Xu, X. Han and X. Bao, J. Phys. 33 W. Limsangkass, P. Praserthdam, S. Phatanasri,
Chem. C, 2008, 112, 5955–5960.
J. Panpranot, N. Poovarawan, W. Jareewatchara, S. Kunjara
D. P. Debecker, M. Stoyanova, U. Rodemerck and
E. M. Gaigneaux, J. Mol. Catal. A: Chem., 2011, 340, 65–76.
K. Gayapan, S. Sripinun, J. Panpranot, P. Praserthdam and
S. Assabumrungrat, RSC Adv., 2018, 8, 11693–11704.
Na Ayudhya and K. Suriye, Catal. Lett., 2014, 144, 1524–1529.
34 S.
Vorakitkanvasin,
W.
Phongsawat,
K.
Suriye,
P. Praserthdam and J. Panpranot, RSC Adv., 2017, 7,
38659–38665.
A. Spamer, T. I. Dube, D. J. Moodley, C. van Schalkwyk and 35 H. Xu, M. Sun, S. Liu, Y. Li, J. Wang and Y. Chen, RSC Adv.,
J. M. Botha, Appl. Catal., A, 2003, 255, 153–167. 2017, 7, 24177–24187.
0 A. G. Basrur, S. R. Patwardhan and S. N. Was, J. Catal., 1991, 36 X.-L. Yang, R. Gao, W.-L. Dai and K. Fan, J. Phys. Chem. C,
27, 86–95. 2008, 112, 3819–3826.
1
1
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 28555–28568 | 28567

View Article Online
RSC Advances
Paper
3
7 G. Chen, M. Dong, J. Li, Z. Wu, G. Wang, Z. Qin, J. Wang and 39 N. Liu, S. Ding, Y. Cui, N. Xue, L. Peng, X. Guo and W. Ding,
W. Fan, Catal. Sci. Technol., 2016, 6, 5515–5525. Chem. Eng. Res. Des., 2013, 91, 573–580.
8 S. Huang, S. Liu, Q. Zhu, X. Zhu, W. Xin, H. Liu, Z. Feng, 40 D. P. Debecker, M. Stoyanova, U. Rodemerck, F. Colbeau-
3
C. Li, S. Xie, Q. Wang and L. Xu, Appl. Catal., A, 2007, 323,
4–103.
Justin, C. Boiss `e re, A. Chaumonnot, A. Bonduelle and
C. Sanchez, Appl. Catal., A, 2014, 470, 458–466.
9
28568 | RSC Adv., 2018, 8, 28555–28568
This journal is © The Royal Society of Chemistry 2018