Effect of direct ultrasound synthesis via a sesquihydrate route on bismuth-promoted vanadyl pyrophosphate catalysts

Article abstract of DOI:10.1002/jccs.201900075

A series of 1, 3, and 5% Bi-doped vanadium phosphate catalyst catalysts were prepared via sesquihydrate route using direct ultrasound method and were denoted as VPSB1, VPSB3, and VPSB5, respectively. These catalysts were synthesized solely using a direct ultrasound technique and calcined in a n-butane/air mixture. This study showed that catalyst synthesis time can be drastically reduced to only 2 hr compared to conventional 32–48 hr. All Bi-doped catalysts exhibited a well-crystallized (VO)₂P₂O₇ phase. In addition, two V⁵⁺ phases, that is, β-VOPO₄ and α_II-VOPO₄, were observed leading to an increase in the average oxidation state of vanadium. All catalysts showed V2p_3/2 at approx. 517 eV, giving the vanadium oxidation state at approx. 4.3–4.6. Field-emission scanning electron microscopy micrographs showed the secondary structure consisting of thin and small plate-like crystal clusters due to the cavitation effect of ultrasound waves. VPSB5 showed the highest amount of oxygen species removed associated with the V⁵⁺ and V⁴⁺ species in temperature-programmed reduction in H₂ analyses. TheX-ray absorption near edge structure (XANES) measurement showed the occurrence of vanadium oxide reductions in hydrogen gas flow, indicating the presence of V⁴⁺ and V⁵⁺ species. Higher average valence states of V⁵⁺, indicating more V⁵⁺ phases, were present. The addition of bismuth has increased the activity and selectivity to maleic anhydride.

Full text of DOI:10.1002/jccs.201900075

Received: 28 February 2019
DOI: 10.1002/jccs.201900075
Revised: 11 May 2019
Accepted: 13 May 2019
A R T I C L E
Effect of direct ultrasound synthesis via a sesquihydrate route
on bismuth-promoted vanadyl pyrophosphate catalysts
Kang-Zhi Goo1 | Yeow-Hong Yap1 | Kuen-Song Lin2 | Loong-Kong Leong1
1Department of Chemical Engineering, Lee
Kong Chian Faculty of Engineering and
Abstract
Science, Universiti Tunku Abdul Rahman,
Kajang, Malaysia
2Department of Chemical Engineering,
Materials Science/Environmental
Technology Research Center, Yuan Ze
University, Taoyuan, Taiwan, ROC
A series of 1, 3, and 5% Bi-doped vanadium phosphate catalyst catalysts were pre-
pared via sesquihydrate route using direct ultrasound method and were denoted as
VPSB1, VPSB3, and VPSB5, respectively. These catalysts were synthesized solely
using a direct ultrasound technique and calcined in a n-butane/air mixture. This
study showed that catalyst synthesis time can be drastically reduced to only 2 hr
compared to conventional 32–48 hr. All Bi-doped catalysts exhibited a well-
crystallized (VO)2P2O7 phase. In addition, two V5+ phases, that is, β-VOPO4 and
αII-VOPO4, were observed leading to an increase in the average oxidation state of
vanadium. All catalysts showed V2p3/2 at approx. 517 eV, giving the vanadium
oxidation state at approx. 4.3–4.6. Field-emission scanning electron microscopy
micrographs showed the secondary structure consisting of thin and small plate-like
crystal clusters due to the cavitation effect of ultrasound waves. VPSB5 showed
the highest amount of oxygen species removed associated with the V5+ and V4+
species in temperature-programmed reduction in H2 analyses. TheX-ray absorption
near edge structure (XANES) measurement showed the occurrence of vanadium
oxide reductions in hydrogen gas flow, indicating the presence of V4+ and V5+
species. Higher average valence states of V5+, indicating more V5+ phases, were
present. The addition of bismuth has increased the activity and selectivity to maleic
anhydride.
Correspondence
Loong-Kong Leong, Department of
Chemical Engineering, Lee Kong Chian
Faculty of Engineering and Science,
Universiti Tunku Abdul Rahman, Jalan
Sungai Long, Bandar Sungai Long, Kajang,
Selangor 43000, Malaysia.
Email: leonglk@utar.edu.my
Funding information
Immortal Green Industrial Sdn. Bhd.
K E Y W O R D S
bismuth, butane oxidation, direct ultrasound synthesis, selective oxidation, sesquihydrate, vanadyl
pyrophosphate catalyst
occurred.[4,5] Vanadium pyrophosphate, (VO)2P2O7, is
known as the active and selective catalyst for the oxidation
1 | INTRODUCTION
The partial oxidation of n-butane over the vanadium phos-
phate catalyst (VPO) is probably the most successful path-
way for the production of maleic anhydride.[1–3] The
oxidation of n-butane to maleic anhydride using a VPO cata-
lyst is a 14 electron oxidation occurring on the catalyst sur-
face. In this process, eight hydrogen atoms were removed,
three oxygen atoms were inserted, and a ring closure
of n-butane to maleic anhydride.
The modification of the VPO catalyst properties by addi-
tion of promoters has been growing popularly especially in
recent years.[1–3,6] Beneficial effects such as structural as
well as electronic changes or enhancement in the activity
and selectivity of the bulk catalyst can be achieved with the
addition of promoters. A wide range of cations such as Ce,
© 2019 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J Chin Chem Soc. 2019;1–9.
http://www.jccs.wiley-vch.de
1

2
GOO ET AL.
Co, Cd, Ni, Zn, Bi, Cu, Li, V, Zr, Mg, Ti, La, Mo, Nb, B,
Fe, and Cr have been reported in patents.[2,3,5] Among the
popular promoters for VPO catalysts, bismuth (Bi) has been
intensively studied for the past two decades.[3,7] Some
researchers reported that the addition of bismuth increases
the activity and selectivity to maleic anhydride.[3,6–9] This
element promotes the catalytic activity through the genera-
tion of surface defects in which the Lewis acid sites are
likely associated, thereby enhancing the activation of the
alkane.[6] Past studies on Bi-doped catalysts prepared via the
VPD method using the conventional preparation route
showed an increment in activity.[6,10]
It is well known that the catalyst preparation method
affects the channel structure, acid site density, oxidation
state, and the location of the vanadium species, and these
factors were found to affect the catalytic performance of n-
butane conversion .[11,12] Moreover, the conventional prepa-
ration time for the VPO catalyst was reported to be around
32–48 hr,[3,8,11–14] therefore alternative methods for generat-
ing catalysts in a shorter duration are highly desirable.[11]
Some developments in the synthesis technique such
as mechanosynthesis, mechanochemical treatment,[13,15]
microwave,[11,16,17] and hydrothermal[18] were reported in
the past. Recently, the use of ultrasound in chemistry has
been discovered to not only enhance the reaction rates but
also direct reactions onto different pathways.[11,12,17,19,20] It
is believed that using ultrasound irradiation can enhance the
chemical reaction and mass transfer via the process of acous-
tic cavitation.[11,21] The ultrasound approach has been used
in the preparation of a wide range of chemical materials such
as ceramics, alloys, composites, and polymers due to its cav-
itation effect on chemical processes.[11,17,20] This cavitation
effect caused by the ultrasound waves, that is the sound pres-
sure waves produced under the ultrasound irradiation fre-
quency from ~20 to 100 MHz, could improve the chemical
reaction and mass transfer via the physicochemical changes
in the medium.[11,20]
FIGURE 1 Powder XRD patterns of the undoped and Bi-doped
catalysts synthesized using the direct ultrasound method
technique are shown in Figure 1. All the catalysts were
found to consist of a mixture of the (VO)2P2O7 phase
(JCPDS File No. 34-1381) and the V5+ phase; i.e., β-VOPO4
(JCPDS File No. 27-0948) and αII-VOPO4 (JCPDS File
No. 34-1247). The diffractogram of the prepared catalysts
showed a similar diffraction pattern comprised of a well-
crystallized (VO)2P2O7 phase with the main characteristic
peaks appearing at 2θ
=
22.9ꢀ, 28.4ꢀ, and 29.9ꢀ,
corresponding to the reflection of (0 2 0), (2 0 4), and (2 2
1) planes, and the additional peaks observed at 2θ = 25.1ꢀ,
27.3ꢀ, and 29.3ꢀ are attributed to the V5+ phase. The peaks
of the V5+ phases were shown to be more intense as the dop-
ing percentage increased, thus indicating that the addition of
bismuth promotes the formation of V5+ phases, which con-
tributes to higher selectivity to maleic anhydride. The exis-
tence of the V5+ phase was also demonstrated in the redox
titration and X-ray photoelectron spectroscopy (XPS) ana-
lyses later in this article, which showed that Bi-doped cata-
lysts had a higher concentration of V5+ and average
oxidation state compared to the undoped catalysts. The Bi-
doped catalysts synthesized using the direct ultrasound tech-
nique showed similar XRD profile data in comparison to the
conventionally prepared Bi-doped catalysts.[3]
The reflection peaks of Bi-doped VPO catalysts at all
three major peaks (2θ = 22.9ꢀ, 28.4ꢀ, and 29.9ꢀ) are shown
to be less intense with broader width sizes as compared to
VPSU. The peak intensities reduced as the amount of doping
increased from low to high percentage. The crystallite sizes
of the (0 2 0) reflection plane for VPSU, VPSB1, VPSB3,
and VPSB5 were calculated as 190.75, 100.14, 79.49, and
125.16 Å respectively, whereas the crystallite sizes of the
(2 0 4) reflection plane catalysts produced were 264.03,
70.63, 51.93, and 92.53 Å, respectively (Table 1). Compared
to the VPSU, all the Bi-doped catalysts showed an obvious
Therefore, in this study, the Bi-doped VPO catalysts were
prepared using the vanadyl hydrogen phosphate
sesquihydrate precursor (VOHPO4Á1.5H2O) via a direct
ultrasound irradiation synthesis process, denoted as the
VPSB route. The effects of different percentages of the bis-
muth promoter on the physicochemical properties, reactivity,
and catalytic performances of the VPO catalysts were
examined.
2 | RESULTS AND DISCUSSION
2.1 | X-ray diffraction
The X-ray diffraction (XRD) patterns of the undoped and
Bi-doped catalysts obtained using direct ultrasound

GOO ET AL.
3
TABLE 1 The X-ray diffraction data
of Bi-doped catalysts
Linewidtha
(0 2 0) (ꢀ)
Linewidthb
(2 0 4) (ꢀ)
Crystallite sizec Crystallite sizec
Catalyst
VPSU
(0 2 0) (Å)
190.75
100.14
79.49
(2 0 4) (Å)
264.03
70.63
0.4200
0.8000
1.0080
0.6400
0.5370
0.4267
0.6767
0.3066
1.1475
1.5614
0.8767
0.6100
0.7080
1.8400
VPSB1
VPSB3
51.93
VPSB5
125.16
149.21
187.78
118.4
92.53
VPSRB1d
VPSRB3d
VPSRB5d
132.79
114.41
44.02
aFWHM of (0 2 0) reflection.
bFWHM of (2 0 4) reflection.
cCrystallite thickness obtained by means of Scherrer's formula.
dConventionally prepared Bi-doped catalysts by Leong et al[3]
reduction in crystallite sizes. Besides, VPSB1 and VPSB3
prepared using the direct ultrasound technique showed
smaller crystallite sizes when compared to the convention-
ally prepared Bi-doped catalysts (1 and 3%). Whereas
VPSB5 produced using the ultrasound method showed
higher crystallite sizes compared to the conventionally pro-
duced Bi-doped 5% catalysts.[3] The crystallinity and sharper
peaks could be observed in this study with the application of
ultrasound. The smaller crystallite sizes allowing a higher
surface area may contribute to the improvement in catalytic
performances. Furthermore, the catalyst production time was
drastically reduced to only 2 hr in the presence of a well-
crystallized (VO)2P2O7 phase .[3,11,14,21,22]
the V5+ phase and the higher average oxidation state with a
higher concentration of V5+
.
As illustrated in Table 2, the V2p3/2 binding energy
values of the VPSB1, VPSB3, and VPSB5 were 517.49,
517.86, and 517.87 eV, respectively, which were approxi-
mate to the value of the undoped catalyst, 517.58 eV. Hence,
it can be suggested that the valence states of vanadium were
in the range of 4–5 and the vanadium species in the synthe-
sized catalysts were mainly in the form of
[4,23–27]
(VO)2P2O7.
As shown in Table 2, the P2p binding
energies detected were almost the same among all of the
samples, which revealed that the element of P was present
on the surface, this phenomenon was good as the slight
excess of phosphorus helps in stabilizing the V4+ oxidation
state.[29,30] As for the binding energies of oxygen, O1s, the
values were very similar (532.0 eV) for all the samples,
which were close to the reference, approx. 531 eV.[4,24–27]
The average oxidation state of vanadium was calculated and
shown in Table 2, the range of which is 4.30 to 4.62.
2.2 | X-ray photoelectron spectroscopy
Figure 2 shows a high-resolution XPS spectrum containing
the O1s, V2p1/2, and V2p3/2 photoelectron peaks for VPSU,
VPSB1, VPSB3, and VPSB5. The V2p3/2 peak could be
separated into two components at 516.9 and 518.0 eV
(Figure 2), which can be related to V4+ and V5+ species,
respectively.[4,17,23–28] The percentages of V4+ and V5+ spe-
cies present were calculated using the deconvolution of the
V2p3/2 peak. The VPSB1 shown to have lowest yields for
the V5+ contributions, revealing that this dopant dosage has
a lower selectivity to the maleic anhydride. While VPSB5
on the other hand showed higher V5+ and V4+ contributions
as compared to the others, which were 72 and 28%, respec-
tively, with a vanadium oxidation state of 4.62. As compared
to the undoped VPSU with 64% of V5+ and 36% of V4+ con-
tributions, the VPSB5 demonstrated higher V5+ contribu-
tion, leading to better selectivity to maleic anhydride.
A higher average oxidation state of vanadium was obtained
due to addition of bismuth that promotes the formation of
the V5+ phase in the catalyst. This is in agreement with the
XRD and redox titration results that showed the presence of
2.3 | Field-emission scanning electron
microscopy
The surface morphologies and particle sizes with different
magnifications of the catalysts obtained from field-emission
scanning electron microscopy (FE-SEM) are shown in
Figure 3. The catalysts showed similar principal and second-
ary structures, consisting of different sizes of plate-like crys-
tals with further agglomeration, forming more compact
structures. However, unlike the conventional synthesized
catalysts that show rosette-shape aggregates,[3,11,14] these
crystallites exhibit much smaller fracturing pieces, stacking
on each other. These agglomerates presumably are the
(VO)2P2O7 platelets that preferentially expose the (1 0 0)
crystal plane.[3,11] Abon et al reported the increase of the
crystallinity of the (VO)2P2O7 phase would change the mor-
phology of the catalyst particles and bring about the

4
GOO ET AL.
catalysts still showed further agglomeration among the clus-
ters with additional layered plate-like crystals at the surface
of the clusters. The smaller and thinner clusters observed
may be due to the cavitation effect of ultrasound during the
synthesis process, which could have contributed to the incre-
ment in the surface area, leading to the increment in the
exposure of the (1 0 0) reflection plane. As shown in
Figure 3, the overall particle sizes of VPSB1, VPSB3, and
VPSB5 were less than 1.0 μm, indicating that smaller crys-
tallites are produced using the ultrasound irradiation tech-
nique. The reduction indicates that there will be increment in
surface areas, which leads to more surface available for cata-
lytic reactions, thus giving higher catalytic activity. The
micrographs were also in good agreement with the results
obtained from the XRD patterns where the increase in reflec-
tion intensity would bring out the formation of more split
[9,13,31]
(VO)2P2O7.
2.4 | Chemical analysis
The composition of vanadium and the percentages of V5+
and V4+ oxidation states are summarized in Table 3. Chem-
ical analysis using inductively coupled plasma-optical
emission spectrometer (ICP-OES) indicated that all Bi-
doped catalysts exhibited a higher phosphorus content with
the increase observed in the P/V atomic ratio from 1.11
(undoped) to 1.23 for VPSB1, 1.25 for VPSB3, and 1.26
for VPSB5. Compared to conventional Bi-doped
catalysts,[3] the increment of the P/V atomic ratio was less
than 0.01 for the same dopant dosage applied, which indi-
cates that a similar composition can be produced using the
ultrasound technique in a shorter duration. The slightly
higher P/V atomic ratio exhibited by the Bi-doped catalysts
in the present study helps in stabilizing the V4+ oxidation
and preventing the oxidation of (VO)2P2O7. The chemical
analyses further confirmed the presence of bismuth only in
the doped catalysts with Bi/V atomic ratios of VPSB1,
VPSB3, and VPSB5 shown to be 0.012, 0.032, and 0.047,
respectively. The average oxidation numbers obtained were
4.34, 4.39, and 4.58 for VPSB1, VPSB3 and VPSB5,
respectively. The average oxidation state slightly higher
than 4.0 is associated with the P/V ratio slightly higher than
1.0, showing the characteristics of the (VO)2P2O7
phase.[3,4,14,21] The redox titration and XPS results are
shown to have the same trend, and in the presence of Bi,
the average oxidation number was found to be higher than
that of the undoped catalyst. The average oxidation number
was observed to be slightly higher than 4.0 due to the pres-
ence of V5+ species (shown in XRD and XPS profiles), as
Bi promotes the formation of V5+ species.
FIGURE 2 XPS spectra obtained with monochromatic AlKα
(1,486.74 eV) excitation with O1s,V2p1/2 and V2p3/2 photoelectron
peaks and fitting curve for; (a) VPSU, (b)VPSB1, (c)VPSB3, and
(d) VPSB5
formation of more split agglomerates.[4] Although the Bi-
doped catalysts synthesized using the direct ultrasound tech-
nique do not agglomerate into the characteristic rosette-
shape as the conventional Bi-doped catalysts,[3] these

GOO ET AL.
5
TABLE 2 XPS results of (VO)2P2O7 catalysts
Binding energy (eV)
Catalyst
VPSU
V2p3/2
517.58
517.49
517.86
517.87
P2p
O1s
Average oxidation state of vanadium (Vox
)
V4+ (%)
36
V5+ (%)
64
133.99
133.61
133.61
133.60
531.97
531.49
531.52
531.40
4.03
4.30
4.53
4.62
VPSB1
VPSB3
VPSB5
61
39
44
56
28
72
catalysts, are listed in Table 4. The Bi-doped catalysts
showed three peak maxima in the reduction by H2
(Figure 5). The first two peaks corresponded to the reduction
of the V5+ phase, while the last peak corresponded to the
removal of lattice oxygen from the active V4+ phase.[3,14,21]
The peak attributed to V4+ is associated with the removal of
O− anions and the peak from V5+ is related to the oxygen
2.5 | X-ray absorption near edge structure
(XANES) measurement of synthesized
vanadium pyrophosphate catalysts
The X-ray absorption edge and pre-edge positions were used
in this work to provide basic information on the different
oxidation states of V (vanadium) in the as-synthesized cata-
lysts to further understand the surface reaction mechanisms.
The normalized V K-edge XANES patterns of the VPSU,
species of O2− [3,4,14,21]
.
The reduction patterns changed significantly with the
addition of bismuth to the VPO catalyst, in which the reduc-
tion of V5+ species appeared to be predominant, whereas the
reduction peak of V4+ was decreased. A similar trend was
also reported[8] in a previous study in which the Bi-doped
catalysts were produced using the reflux conventional
method[3] and the Bi-Fe catalyst was produced using the
reflux VPD method.
VPSB1, VPSB3, and VPSB5 were determined using V3+
,
V4+, and V5+ standards as presented in Figure 4. Three
peaks above the absorption edge in the region 0–50 eV were
used to obtain a better fitting of the absorption edge position
in the spectra of both V standards and the as-synthesized cat-
alysts. Based on these adsorption spectra, the pre-edge
absorption positions of the V3+, V4+, and V5+ standards due
to the 1 s-to-3d transition were determined to be 5,468.2,
5,469.1, and 5,470.7 eV, respectively. In addition, the inflec-
tion point of the V K-edge absorption, for oxidation states of
3+, 4+, and 5+, were 5,477.8, 5,479.6, and 5,482.2 eV,
respectively. The pre-edge XANES spectra of V for the as-
synthesized catalysts exhibited an absorbance feature at the
1s-to-3d transition. As for the fresh catalyst samples,
the sharp features were observed at 5,480–5,482 eV
indicating that the fresh catalysts hold an oxidation state of
V4+ or V5+, and not V3+. Due to the occurrence of
vanadium oxide reduction in hydrogen gas flow, the shoul-
ders at 5,490–5,494 eV were attributed to the dipole
allowed, 1s-to-4pxy transition, which indicated the presence
of V4+ and V5+ species in the as-synthesized catalysts.
These adsorption spectra could be fitted well without any
In comparison to the undoped catalyst, the reduction
peaks occurred at lower temperature for the Bi-doped cata-
lyst. First and second peaks of VPSB1 occurred at 790 and
819 K while the last peak appeared at 1138 K with the
amount of oxygen removed from V5+ was significantly
increased to 7.57 × 1020 and the amount of oxygen removed
from V4+ was 4.54 × 1020 atom/g, respectively. The ratio of
oxygen atoms removed from V5+/V4+ was 1.67. Similarly,
VPSB3 gave three peaks appearing at 788, 850, and 1131 K.
The reduction peaks generally shifted to a lower temperature
with an increase of Bi concentration. Total oxygen species
removed from V5+ species was reduced to
1.99 × 1020 atom/g, whereby the oxygen removed from V4+
species was also decreased to 2.54 × 1020 atom/g. The phe-
nomenon of reduction of reactivity in bismuth 3 mol% was
also reported by Leong et al using a conventional prepara-
tion method of the Bi-doped catalyst.[3] The first two peaks
of VPSB5 appeared at slightly higher temperature than
VPSB1 and VPSB3, that is, 856 and 1149 K with the third
peak occurring at 1195 K. The removal of oxygen species
linked to V5+ and V4+ species have tremendously increased
to 2.54 × 1021 atom/g and 1.49 × 1021 atom/g, respectively.
As a result, the ratios of oxygen species removed were
increased. Hence, among the Bi-doped catalysts produced
via the ultrasound technique, VPSB5 probably demonstrated
the highest reactivity.
contribution from V3+
.
2.6 | Temperature-programmed reduction
in H2/N2
Temperature-programmed reduction (TPR) in H2 analyses
were used to investigate the redox properties, the nature and
the oxidising species available from the catalysts. The TPR
results are shown in Figure 5, the total amount of oxygen
removed from each peak, including the peak maxima tem-
peratures and the derived reduction activation energy of the

6
GOO ET AL.
The Bi-doped catalysts produced using the direct ultra-
sound method in general exhibited the reduction peaks at a
lower temperature, thus leading to more labile oxygen,
which consequently resulted in a higher removal of oxygen
species that was attributed to the activity of the catalyst.
Besides, the results shown by TPR in H2 analyses also indi-
cated that the introduction of increasing concentration of Bi
promoter would lead to the increment of the amount of oxy-
gen species associated with V5+ species with slight impact.
These results were in agreement with the results of XRD and
XPS, which also showed the presence of higher amounts of
V5+
.
2.7 | Selective oxidation of n-butane to maleic
anhydride
The catalytic performances of VPO catalysts for oxidation of
n-butane to maleic anhydride were tested at a typical operat-
ing temperature of 673 K (Table 5). The undoped catalyst
synthesized using the direct ultrasound method showed a
conversion of n-butane of 58%, the doped catalysts have
shown a conversion of n-butane of 45, 50, and 65% for
VPSB1, VPSB3, and VPSB5, respectively. As compared to
the conventionally prepared Bi-doped catalysts,[3] there was
an increment observed in the catalytic activity. However,
compared to the undoped catalyst produced using the direct
ultrasound method, addition of higher amounts of bismuth
better improves the catalytic activity. The VPSB1 and the
undoped catalyst showed the same maleic anhydride selec-
tivity, 68%, respectively. Further addition of bismuth had
increased the catalyst selectivity to 73 and 80% for VPSB3
and VPSB5, respectively. These observations indicated that
the introduction of a higher amount of Bi into the catalysts
improved catalyst selectivity toward maleic anhydride.
3 | EXPERIMENTAL
3.1 | Catalyst preparation
The catalysts were prepared via sesquihydrate route and
direct ultrasound method, wherein V2O5 (15 g) was mixed
with aqueous 85% ortho-phosphoric acid (90 mL) and dis-
tilled water (360 mL). The mixture was subjected to direct
ultrasound irradiation (40% amplitude) for 1 hr by an ultra-
sonic probe (2.0 cm diameter) using a 750 W, 20 kHz Ultra-
sonic Power Supply from SnapTec NexTgen (Model
10000-03). The power supply features an automatic ampli-
tude and frequency control circuitry that eliminates the need
for constant adjustments and assures optimum cavitation at
any power level. The tip of the ultrasonic probe was
immersed ꢁ2 cm below the surface of the mixture
(Figure 6) and the catalyst remained suspended in the
FIGURE 3 FE-SEM micrographs of (a) VPSU, (b) VPSB1,
(c) VPSB3, and (d) VPSB5 catalysts

GOO ET AL.
7
TABLE 3 Chemical compositions, average oxidation numbers, and percentages of V4+ and V5+ oxidation states present in doped VPO
catalysts
Inductively coupled plasma
Catalyst
VPSU
Energy-dispersive X-ray
P/V
1.11
1.23
1.25
1.26
Bi/V
V4+ (%)
72.22
65.57
60.89
42.25
V5+ (%)
27.78
34.43
39.11
57.75
Average oxidation number
1.66
1.34
1.52
1.62
—
4.2778
4.3443
4.3911
4.5775
VPSB1
VPSB3
VPSB5
0.012
0.032
0.047
irradiation for 1 hr with 1-butanol (50 mL) and bismuth
nitrate, Bi(NO3)3Á2H2O as the salt of the promoter. At this
stage, bismuth was added according to the mole percentage
of the dopant (1, 3 and 5%). The resulting light blue solid
(VOHPO4Á1.5H2O precursor) was recovered by centrifuga-
tion and washed sparingly with acetone, further oven dried
at 353 K overnight. The obtained precursors were calcined
under a flow of 0.75% n-butane in air mixtures at 673.15 K
(400ꢀC) for 24 hr to produce the active catalysts,
(VO)2P2O7. The activated Bi-doped (1, 3 and 5%) catalysts
were denoted as VPSB1, VPSB3, and VPSB5, respectively.
The same method was used to prepare the undoped catalyst,
denoted as VPSU.
FIGURE 4 The V K–edge-normalized XANES of VPSU,
VPSB1, VPSB3, and VPSB5 catalysts
3.2 | Catalyst characterization
The XRD analyses were carried out using a Shimadzu dif-
fractometer model XRD 6000 employing Cu Kα radiation to
generate diffraction patterns from powder crystalline sam-
ples at ambient temperature. The estimation of the crystallite
size based on X-ray peak broadening is carried out using the
Debye–Scherrer equation. The FWHM of the (0 2 0) and
(2 0 4) planes were used to determine the crystallite size of
the catalysts. The parameter used to determine the crystal
size of the catalysts is the half width of the (0 2 0) reflection.
The thickness of the (2 0 4) plane is indicative of the mean
“length” at the (2 0 4) face, while the thickness of (0 2 0) is
representative of the actual thickness.[2,3,6,14,21] Therefore,
the line width increases with the decrease in the crystallite
size. The chemical composition on the surface of the catalyst
was determined using XPS (Physical Electronic ESCA PHI
1600) at an excitation energy of Al Kα (1,486.6 eV). The
C1s (284.5 eV) peak was used as the calibration standard for
the wide-region spectra of the vanadium and phosphorus
species. The XPS signals of the above species were recorded
with a cylindrical mirror analyzer. In the present practice,
the peak deconvolution procedure where VOPO4 (oxidation
state, +5) with phase structure confirmed by XRD results
provided the reference value for the binding energy of the
V2p3/2 peak. The measured V2p3/2 peak (FWHM < 1.3 eV)
could be well fitted with a single Gaussian with a maximum
FIGURE 5 TPR in H2 profiles of undoped and Bi-doped
catalysts
absence of any external mixing due to the acoustic effect of
ultrasound waves. The brownish solid solution (color of
V2O5) gradually changed to yellow. The solid
(VOPO4Á2H2O) was recovered by centrifugation and was
oven dried at 353 K overnight. Next, the well-synthesized
VOPO4Á2H2O (10 g) was subjected to direct ultrasound

8
GOO ET AL.
TABLE 4 Total number of oxygen atoms removed, values of reduction activation energy, and the ratio of oxygen removal from V5+/V4+ and
V4+/V5+ obtained by the reduction in H2/N2 for Bi-doped catalysts
Reduction activation
energy, Er (kJ/mol)
Ratio of oxygen
Catalyst
Tonset (K)
Peak
Tmax (K)
981
Oxygen removed (atom/g)
4.69 × 1020
1.78 × 1021
2.25 × 1021
1.58 × 1020
5.99 × 1020
4.54 × 1020
1.21 × 1021
1.05 × 1020
9.37 × 1019
2.54 × 1020
4.53 × 1020
1.18 × 1021
1.36 × 1021
1.49 × 1021
4.03 × 1021
removal from V5+/V4+
VPSU
752
1
2
408
482
0.26
1158
Total oxygen removed:
VPSB1
VPSB3
VPSB5
593
1
2
3
790
329
341
473
1.67
819
1138
Total oxygen removed:
669
1
2
3
788
328
354
470
0.79
1.71
850
1131
Total oxygen removed:
672
1
2
3
856
356
478
497
1149
1195
Total oxygen removed:
TABLE 5 Catalytic performances of undoped and doped VPO
catalysts
bulk chemical composition was determined using a sequen-
tial scanning ICP-OES (Perkin Elmer Emission Spectrome-
ter model Plasma 1000). The average oxidation states of
vanadium in all the catalysts were determined by redox titra-
tion following the method of Niwa and Murakami.[32] The
XANES spectra were collected on the Wiggler beam line
16A1 at the National Synchrotron Radiation and Research Cen-
ter in Taiwan. Data were collected in the transmission mode
using a Lytle detector in the regions of V K-edges at room tem-
perature. The data were subsequently normalized using Athena
(vi) software. TPR in H2 analysis was done using a Thermo
Electron TPDRO 1100 apparatus utilizing a thermal conductiv-
ity detector. The catalyst was pretreated in a flow of 20 cm3/
min purified nitrogen, held for 10 mins before cooling to ambi-
ent temperature. Sample analysis was done at a flow of 5.5%
hydrogen in nitrogen (25 cm3/min), starting from room temper-
ature to 1000ꢀC with a heating ramp of 5ꢀC/min.
Catalyst selectivity (%)
Catalyst
VPSU
n-butane conversion (%)
MA
68
CO/CO2
58
45
50
65
32
32
27
20
VPSB1
VPSB3
VPSB5
68
73
80
Ultrasound
Probe
3.3 | Selective oxidation of n-butane to maleic
anhydride
FIGURE 6 The tip of the ultrasonic probe was immersed ꢁ2 cm
The oxidation of n-butane was carried out in a fixed-bed
microreactor at 673 K with gas hourly space velocity
(GHSV) = 2,400 hr−1 with a standard mass of the catalyst
(250 mg). A mixture of 1.0% n-butane and air was fed to the
reactor via a calibrated mass flow controller. The products
were then fed via heated lines to an on-line gas chromatogra-
phy for analysis. The reactor comprised a stainless-steel tube
below the surface of the mixture
at 518 eV.[4,23–27] The morphologies, microstructures, and
particle sizes of the catalysts were investigated using a FE-
SEM (Hitachi, S-4700 Type II). Energy-dispersive X-ray
(EDX) analyses were carried out using EDAX software. The

GOO ET AL.
9
[9] Y. H. Taufiq-Yap, Y. Kamiya, K. P. Tan, J. Nat. Gas Chem.
2006, 15, 297.
[10] Y. H. Taufiq-yap, K. G. Chee, K. C. Waugh, Y. Kamiya, React.
Kinet. Catal. Lett. 2005, 84, 271.
[11] U. Pillai, E. Sahle-demessie, R. S. Varma, Appl. Catal. A. Gen.
2003, 252, 1.
with the catalyst held in place by plugs of quartz wool. A
thermocouple was located in the center of the catalyst bed
and temperature control was typically 1 K. Carbon mass
balances of ≥95% were typically observed.
[12] C. W. Yee, Y. H. Taufiq-Yap, Y. C. Wong, Y. H. Taufiq-Yap,
Adv. Mat. Res. 2011, 364, 25.
4 | CONCLUSIONS
[13] Y. H. Taufiq-Yap, Y. C. Wong, Y. Kamiya, W. J. Tang, J. Nat.
Gas Chem. 2008, 17, 232.
[14] Y. H. Taufiq-Yap, L. K. Leong, M. Z. Hussein, R. Irmawati,
S. B. Abd Hamid, Catal. Today 2004, 93–95, 715.
[15] Y. H. Taufiq-Yap, Y. C. Wong, Z. Zainal, M. Z. Hussein, J. Nucl.
Relat. Technol. 2009, 6, 1.
[16] L. Zeng, H. Jiang, J. Niu, J. Mol. Catal. A Chem. 2005, 232, 119.
[17] Y. H. Taufiq-Yap, J. R. Joshua Hoh, Y. C. Wong, J. Appl. Sci.
2011, 11, 2370.
[18] Y. H. Taufiq-Yap, A. R. M. Hasbi, M. Z. Hussein,
G. J. Hutchings, J. Bartley, N. Dummer, Catal. Lett. 2006,
106, 177.
[19] T. J. Mason, Ultrasonics 1986, 24, 245.
[20] B. Cos¸kuner, A. Kantürk Figen, S. Pis¸kin, J. Chem. 2014,
2014, 1.
[21] Y. C. Wong, Y. H. Taufiq-Yap, Asian J. Chem. 2011, 23, 3853.
[22] S. Ieda, S. Phiyanalinmat, S. I. Komai, T. Hattori, A. Satsuma,
J. Catal. 2005, 236, 304.
[23] K. Gaku, T. Okuhara, M. Makoto, Catal. Lett. 1995, 32, 205.
[24] Y. Suchorski, B. Munder, S. Becker, L. Rihko-Struckmann,
K. Sundmacher, H. Weiss, Appl. Surf. Sci. 2007, 253, 5904.
[25] E. Kleimenov, H. Bluhm, M. Hävecker, A. Knop-Gericke,
A. Pestryakov, D. Teschner, J. A. Lopez-Sanchez, J. K. Bartley,
G. J. Hutchings, R. Schlögl, Surf. Sci. 2005, 575, 181.
[26] H. Wu, P. Jin, Y. Sun, M. Yang, C. Huang, W.-Z. Weng, et al.,
J. Mol. Catal. A Chem. 2016, 414, 1.
[27] R. W. Wenig, Vanadium-Phosphorous-Oxygen Industrial Cata-
lysts for C4 Hydrocarbon Selective Oxidation to Maleic Anhy-
dride 1986. Retrospective Thesis and Dissertations (Paper 8321),
https://lib.dr.iastate.edu/cgi/viewcontent.cgi?referer=https://www.
google.com/&httpsredir=1&article=9320&context=rtd (accessed:
July 2016).
It is shown in this study that all the bismuth-promoted
(VO)2P2O7 phase produced via direct ultrasound synthesis
contains higher V5+ with a smaller particle size, thereby
indicating better catalytic performances with higher selectiv-
ity to maleic anhydride. The ultrasound irradiation technique
further improved the catalytic performances by increasing
the surface area with better topographical morphologies. The
Bi-doped catalysts synthesized using the direct ultrasound
technique were shown to be of crystals having thinner and
smaller plate-like structures. The V2p3/2 binding energy ana-
lyses further concreted the findings of the diffractogram of
XRD, in which the Bi-doped catalysts exhibited good crys-
tallinity with characteristic peaks of the (VO)2P2O7 catalysts.
The preparation of Bi-doped catalysts through the ultrasound
synthesis technique has drastically reduced the synthesis
time to only
2 hr compared to the conventional
32–48 hr.[3,8,11–14]
ACKNOWLEDGMENTS
Financial support from Immortal Green Industrial Sdn. Bhd.
and Yuan Ze University, Taiwan was gratefully
acknowledged.
ORCID
Loong-Kong Leong https://orcid.org/0000-0001-7113-
299X
[28] S. Luciani, F. Cavani, V. Dal Santo, N. Dimitratos, M. Rossi,
C. L. Bianchi, Catal. Today 2011, 169, 200.
REFERENCES
[29] L. O'Mahony, D. Zemlyanov, M. E. Smith, B. K. Hodnett, Appl.
Catal. A. Gen. 2003, 251, 327.
[30] J. Guan, H. Xu, S. Jing, S. Wu, Y. Ma, Y. Shao, Q. Kan, Catal.
Commun. 2008, 10, 276.
[31] Y. H. Taufiq-Yap, C. S. Saw, Catal. Today 2008, 131, 285.
[32] M. Niwa, Y. Murakami, J. Catal. 1982, 76, 9.
[1] V. Guliants, J. B. Benziger, S. Sundaresan, I. E. Wachs,
A. M. Hirt, Catal. Lett. 1999, 62, 87.
[2] G. J. Hutchings, R. Higgins, J. Catal. 1996, 162, 153.
[3] L. K. Leong, K. S. Chin, Y. H. Taufiq-Yap, Catal. Today 2011,
164, 341.
[4] M. Abon, J.-C. Volta, Appl. Catal. A. Gen. 1997, 157, 173.
[5] R. L. AL-Otaibi, Preparation and Characterisation of Vanadium
Phosphorus Oxide Catalysts for Butane Oxidation to Maleic
Anhydride, Cardiff Universty, Wales 2010.
[6] A. A. Rownaghi, Y. H. Taufiq-Yap, F. Rezaei, Catal. Lett. 2009,
130, 504.
[7] Y. H. Taufiq-Yap, C. S. Yuen, N.@. M. N. Suziana, R. Irmawati,
Chin. J. Catal. 2014, 35, 270.
[8] C. K. Goh, Y. H. Taufiq-Yap, G. J. Hutchings, N. Dummer,
J. Bartley, Catal. Today 2008, 131, 408.
How to cite this article: Goo K-Z, Yap Y-H,
Lin K-S, Leong L-K. Effect of direct ultrasound
synthesis via a sesquihydrate route on bismuth-
promoted vanadyl pyrophosphate catalysts. J Chin
Chem Soc. 2019;1–9. https://doi.org/10.1002/jccs.
201900075