Oxidative dehydrogenation of n-butane to butadiene catalyzed by new mesoporous mixed oxides NiO-(beta-Bi2O3)-Bi2SiO5/SBA-15 system

Article abstract of DOI:10.1016/j.mcat.2020.110893

NiO-beta-Bi₂O₃-Bi₂SiO₅/SBA-15 catalysts, containing 10?20 wt% Ni and 10?30 wt% Bi loaded as metal weight on mesoporous SiO₂ support (SBA-15), were utilized for the oxidative dehydrogenation of n-butane to butadiene, comparing from the viewpoint of Bi oxide phase and combination balance with Ni oxide and the support. Bi₂SiO₅ and beta-Bi₂O₃ phases change depending on Ni/Bi ratio. “Reverse core-shell” structure with uniform mesoporosity catalytically active for the oxidative dehydrogenation of n-butane to butadiene was successfully synthesized with semi-dry conversion method. The shell was formed as Bi₂SiO₅/SBA-15 by partially dissolving an array of silanol inside SBA-15 mesopores with impregnated bismuth nitrate. The Bi₂SiO₅ of the shell was increased by co-impregnated nickel nitrate. The layered core faced to mesopore was formed as catalytically active NiO/beta-Bi₂O₃ on the shell. The degrees of formation Bi₂SiO₅ and beta-Bi₂O₃ reflected in the butadiene selectivity through changing the reducibility and dispersion properties. The catalyst with moderate loading of 20 wt% Ni and 10 wt% Bi exhibited a high advantage in the n-butane conversion: 30 % and butadiene selectivity: 49 % at 450 °C compared to the less and excess loaded catalysts. The catalytic performance of other mesoporous silica support catalysts also changed differently depending on Bi₂SiO₅ and beta-Bi₂O₃ phases.

Full text of DOI:10.1016/j.mcat.2020.110893

MolecularCatalysis488(2020)110893
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Oxidative dehydrogenation of n-butane to butadiene catalyzed by new
mesoporous mixed oxides NiO-(beta-Bi2O3)-Bi2SiO5/SBA-15 system
G. Tanimua,b,*, A.M. Aitanib, S. Asaokab, H. Alasiria,b
a Chemical Engineering Department, KFUPM, Dhahran, 31261, Saudi Arabia
b Center for Refining & Petrochemicals, Research Institute, KFUPM, Dhahran, 31261, Saudi Arabia
A R T I C L E I N F O
A B S T R A C T
Keywords:
n-Butane
NiO-beta-Bi2O3-Bi2SiO5/SBA-15 catalysts, containing 10−20 wt% Ni and 10−30 wt% Bi loaded as metal weight
on mesoporous SiO2 support (SBA-15), were utilized for the oxidative dehydrogenation of n-butane to butadiene,
comparing from the viewpoint of Bi oxide phase and combination balance with Ni oxide and the support. Bi2SiO5
and beta-Bi2O3 phases change depending on Ni/Bi ratio. “Reverse core-shell” structure with uniform meso-
porosity catalytically active for the oxidative dehydrogenation of n-butane to butadiene was successfully syn-
thesized with semi-dry conversion method. The shell was formed as Bi2SiO5/SBA-15 by partially dissolving an
array of silanol inside SBA-15 mesopores with impregnated bismuth nitrate. The Bi2SiO5 of the shell was in-
creased by co-impregnated nickel nitrate. The layered core faced to mesopore was formed as catalytically active
NiO/beta-Bi2O3 on the shell. The degrees of formation Bi2SiO5 and beta-Bi2O3 reflected in the butadiene se-
lectivity through changing the reducibility and dispersion properties. The catalyst with moderate loading of
20 wt% Ni and 10 wt% Bi exhibited a high advantage in the n-butane conversion: 30 % and butadiene selectivity:
49 % at 450 °C compared to the less and excess loaded catalysts. The catalytic performance of other mesoporous
silica support catalysts also changed differently depending on Bi2SiO5 and beta-Bi2O3 phases.
Dehydrogenation
Butadiene
SBA-15
Bi2SiO5
1. Introduction
dehydrogenations were focused on. Our prior studies [16–19] utilized
nanoparticles of binary oxide Ni-Bi-O species deposited on commer-
The global demand for butadiene in the manufacturing of synthetic
rubber was about 12.3 million tons in 2018 and it is expected to grow
by 4.5 % annually during the next five years [1]. On-purpose produc-
tion methods have been investigated to supplement the decrease in
butadiene supply due to the shift from naphtha to ethane crackers.
Oxidative dehydrogenation (ODH) reaction is one of the promising
methods aimed at bridging the gap between the growing demand and
supply of butadiene. Many researchers have investigated n-butane
oxidative dehydrogenation to butenes/butadiene mainly using vana-
dium based catalysts composed of V2O5/SiO2 [2], V2O5/silica gel [3],
V/supports (TiO2, ZrO2 and Al2O3) [4,5], V/TiO2-SiO2 [6], Mg3(VO4)2
species over Al2O3, ZrO2, MgO and CeO2 [7,8], VOx/supports (USY,
NaY, γ-Al2O3, α-Al2O3) [9], VOx/SBA-15 [10], V/supports (HMS, SBA-
16, SBA-15, MCM-48) [11], VOx/Ti-HMS [12], V-MCM-41 [13], V/Ti-
SBA-15 [14], and Mo-V-MgO [15]. On the other hand, Bi-Mo-O based
catalysts are well-known for ODH of n-butenes to butadiene. Referring
to these literatures, the functions of the catalysts in the first (butane to
1-butene/2-butenes) and second (butenes to 1,3-butadiene) step
cially available Al2O3, SiO2, and ZrO2 for the ODH of n-butane with
high selectivity to butadiene. The contribution of Bi oxide without Mo
in stabilizing the active site NiO nanoparticles selective to butadiene
[16], the role of calcination (two steps of 350 and 590 °C in air) in
controlling the Bi oxide beta-Bi2O3 phase formations [17], and the sy-
nergetic effects of the active site NiO modification by partial substitu-
tion with Co and Fe [18], have been reported. The role of support in
controlling the redox character of the active species, metal-support
interaction and acid-base properties have been highlighted [19]. In an
advanced studies on the effect of supports, an enhanced performance
was obtained with SiO2 sol supported catalyst compared to the un-
supported, Al2O3 and ZrO2 gel supported catalysts or the sol supported
ones.
Mesoporous silica support shows enhanced performance in catalytic
applications through stabilizing the performance of the impregnated
metal oxides nanoparticles with a uniform dispersion. This is mainly
due to their ordered structure, high surface area, high chemical and
thermal stabilities, and well-defined pore sizes [20–23]. Mesoporous
⁎ Corresponding author at: Center for Refining & Petrochemicals, P.O. Box 5040, KFUPM, Dhahran, 31261, Saudi Arabia.
E-mail addresses: gtu4ever@yahoo.com (G. Tanimu), maitani@kfupm.edu.sa (A.M. Aitani), asataro4322@gmail.com (S. Asaoka),
alasiri@kfupm.edu.sa (H. Alasiri).
https://doi.org/10.1016/j.mcat.2020.110893
Received 20 June 2019; Received in revised form 21 February 2020; Accepted 14 March 2020
2468-8231/©2020ElsevierB.V.Allrightsreserved.

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
siliceous carriers (SBA-15, silica foam and MCM-41) are utilized in the
synthesis of highly dispersed tungsten oxide catalyst for butene me-
tathesis reaction [24–26]. Nanoparticles of vanadium deposited on
mesoporous silica supports have been utilized for the ODH of n-butane
[10–14]. Supports of highly ordered mesoporous silica (SBA-15, silica
foam, and MCM-41) loaded with 20 wt% Ni and 30 wt% Bi, were also
utilized in our previous study [27]. The binary-oxide impregnated
catalysts were probed for n-butane ODH to butadiene and compared
with traditional silica gel catalysts. Mesoporous silicas as catalyst sup-
port are superior to the traditional silica gels, where the SBA-15 me-
soporous silica supported catalyst gave the best performance [27]. The
interaction of BiOx species with microporous and mesoporous silicon
oxides was reported by Dumitru et al. [28]. The BiOx species were in-
corporated into the ZSM-5 pores during hydrothermal synthesis and it
was concluded that the well-dispersed and occluded BiOx clusters are
effective for the liquid-phase oxidation of hydrocarbons. The strength
and/or type of acid sites present and the redox properties of the catalyst
were controlled by the BiOx species and its interaction with the support
[28]. This was extended in a series of papers by different authors, for
mainly oxidation and photocatalytic studies [29–35].
reactants flow rate was kept at 31.2 ml/min. The performance of the
catalysts were evaluated at 450 °C and Oxygen to n-butane ratio
=2.0 mol/mol. Analyses of products was achieved with an online gas
chromatograph (GC) system (Agilent, 7890 N). Flame ionization de-
tector (FID) with GC-Gas Pro capillary column was utilized in oxyge-
nates and hydrocarbons identification. All the gases were identified
with a Thermal conductivity detector (TCD) having He and Ar as car-
riers for the TCD columns. Standard samples were used for confirming
the products. The conversion of n-butane and the selectivities of the
products were obtained using based on carbon mass balance.
2.3. Characterization of catalysts
The physical properties were obtained using N2 adsorption-deso-
rption isotherm with Micromeritics ASAP 2020 instrument, Norcross,
GA. BJH adsorption method was used in obtaining the pore surface
area, pore volume and average pore diameter. X-ray diffraction (XRD)
patterns of the calcined catalysts were recorded from 5° to 90° dif-
fraction angle using X-ray diffractometer (Rigaku Miniflex II) utilizing
Cu Kα radiation at λ = 1.5406 Å and 30 mA and 40 kV operating
parameters, at a speed of 0.5°/min and step size of 0.02°. High
Resolution Transmission Electron Microscopy (HRTEM), JEM-2100 F
model having an acceleration voltage of 200 kV was used in analyzing
the morphologies of the catalysts. Temperature Programmed Reduction
(TPR) was used in determining the catalysts reducibility. This was
achieved with a chemisorption instrument [27]. H2/Ar (5 vol%/95 vol
%) gas mixture at a flow rate of 50 cm3/min was used in the TPR
measurement. 100 mg of the calcined catalyst was preheated for 3 h at
300 °C under inert He and then cooled to room temperature. It was then
increased up to 1000 °C at the rate of 20 °C /min. The intake of H2 was
measured with TCD and CuO was applied as a reference for H2 con-
sumption calibration. The catalysts binding energy and bonding states
were analyzed using X-ray photoelectron spectroscopy (XPS) with a PHI
5000 Versa Probe II, ULVAC-PHI Inc. spectroscope. The samples were
disc-pelletized and put under a high vacuum prior to the XPS mea-
surement.
This manuscript reports the effect of cooperating BiOx species with
SBA-15 mesoporous silica support on the redox property of the active Ni
species and on the overall catalytic performance during n-butane oxi-
dative dehydrogenation. The metal oxides-support cooperation was
clearly studied using X-ray diffraction, X-ray photoelectron spectro-
scopy and temperature programmed reduction. Other mesoporous silica
supports including silica foam, silica sol, and the traditional silica gel
catalysts have also been utilized in this study as references.
2. Experimental
2.1. Catalyst preparation
SBA-15 mesoporous silica carrier was prepared using tri-block co-
polymer as a structure directing agent. 4 g of Pluronic P123 was in-
troduced into 30 ml of de-ionized water. Stirring continued until clear
solution was achieved. 70 g of 0.28 M HCl was added to the solution
with continuous stirring for 2 h. 9 g of Tetraethyl orthosilicate (TEOS)
was introduced and stirring continued for 24 h at 40 °C and finally
heated for 48 h at 100 °C. Filtration was utilized in recovering the solid
product, de-ionized water was used several times for washing the re-
sidue and then dried at 100 °C overnight. Calcination was done at
550 °C for 6 h for template removal [36]. The other mesoporous silicas
used were silica gel Q10; procured from Fuji Silysia Chemicals Limited,
Japan. MCM-41 was prepared following the method reported by Palani
et al. [37]. Silica foam was prepared using the method obtained in the
literature [38], and silica sol was purchased from Sigma Aldrich.
Co-impregnation technique was utilized in synthesizing all the
supported catalysts to obtain enhanced cohabitation of mixed oxide
nanoparticles [39]. Ni and Bi precursors utilized were respectively Ni
(NO3)2·6H2O (99 %, Fisher Scientific) and Bi(NO3)3·5H2O (98 %, Fluka
Garantie). In a typical synthesis, 990 mg of Ni precursor was dissolved
in 80 ml of de-ionized water. 700 mg of Bi precursor was introduced
after dissolution with continuous stirring. 1000 mg of the support was
then introduced. The mixture was left overnight for Bi species equili-
brium adsorption and followed by evaporative drying at 80 °C for en-
forced deposition of Ni species. The product was dried further for 3 h at
120 °C and two steps calcination at 350 °C for 1 h and 590 °C for 2 h at
the rates of 10 °C/min and 15 °C/min, respectively, was carried out.
3. Results and discussion
3.1. Catalyst property
3.1.1. Surface area and porosity
The dispersion of active sites is mainly determined using the BET
surface area and catalysts pore structure [40]. The surface areas cal-
culated using the BET equation in the linear region of the N2 adsorp-
tion-desorption isotherms (P/Po = 0.05-0.3), all the catalysts pore
properties calculated using the BJH method, are presented together
with the support values in Table 1. A decrease in surface area was
observed for all the catalysts compared to that of the support. The same
trend was also observed in the catalysts pore structure.
The N2 adsorption-desorption isotherm of the 20 Ni-30 Bi-O/SBA-15
catalyst (normalized to support weight) and SBA-15 support is pre-
sented in Fig. 1(a) while the pore size distribution (normalized to
support weight) using the adsorption branch isotherm is shown in
Fig. 1(b). The isotherms are typical of type IV of the IUPAC classifica-
tion due to the existence of the type H1 hysteresis loop, which is
common for mesoporous materials having a well-defined cylindrical-
like pore channel [40]. Adsorption and desorption paths of the isotherm
coincide at low relative pressure (up to P/Po ≈ 0.5) which is an in-
dication that there exists monolayer-multilayer adsorption [41]. The
beginning of the hysteresis loop signifies capillary condensation within
the pores, and its end corresponds to the filling of the pores. Metal
loading resulted in the decrease in the height of the loop due to pore
volume decrease caused by the introduction of metal species within the
support mesopore [42]. The pore size distribution gave a sharp peak for
both the support and the catalyst, averaged around 8−10 nm, which
2.2. Catalytic testing
ODH reaction of n-butane was conducted in an automated fixed-bed
reactor obtained from Microtrac Bel Company, Japan. 300 mg of the
synthesized catalyst was introduced into the reactor to form the catalyst
bed. The contact time of n-butane was fixed at 0.42 h·g/mol. Total
2

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
Table 1
Catalysts and supports physical properties.
Catalyst:20 wt% Ni-30 wt% Bi-O/
BET surface area
Pore surface area
Pore volume
Average pore
diameter
support (Support)
[m2/g-catalyst]a [m2/g-support]b [m2/g-catalyst]c [m2/g-support]d [cm3/g-catalyst]e [cm3/g-
[nm]g
support]f
CSG
158
269
388
91
251
171
309
397
87
272
0.71
0.33
1.55
0.64
1.13
16.8
(18.6)
4.2
(SiO2 gel)
CSB
(242)
427
(263)
491
(1.22)
0.52
(SBA-15)
CSF
(657)
616
(1080)
630
(1.08)
2.46
(4.1)
15.6
(16.4)
29.1
(SiO2 foam)
CSS (SiO2 sol)
(540)
144
(554)
138
(2.27)
1.02
aBET surface area,c,e,gSurface area, pore volume and average pore diameter measured using BJH isotherm, b,d,fSurface area and pore volume calculated to support
weight base by using the equation: SA or PV×[©(MOx/M)+100]/100, where M = metal wt%; MOx= metal oxide wt%; SA = surface area; PV = pore volume.
indicate a narrow mesopore size distribution, typical of mesoporous
SBA-15 material [42,43].
Table 2
Pore size distribution of Ni-Bi-O/SBA-15 catalyst and support.
The pore size distribution was divided into three regions to carefully
examine the effect of the metal loading on the support. The pore surface
area and pore volume of the support and catalyst are presented in
Table 2. The pore surface area and the pore volume of the SBA-15
catalysts decreased to almost 50 % of the support. The main decrease is
observed in the region of smaller pore diameter (1.7–3.9 nm). An ap-
proximately 1.1 nm thickness of the main pore inside surface of the
support has been covered by the metal oxides impregnation. This is an
indication that the Ni and Bi oxide species have been loaded uniformly
into both the smaller pores and main mesopores of SBA-15.
The surface area, pore volume and the average pore diameter of the
catalysts with reduced Ni and/or Bi loading can be found in the sup-
plementary material [S1]. The catalysts exhibited a decrease in surface
area and pore volumes relative to the supports. Average pore diameter
remained unchanged in all the catalysts.
Material
Support: SBA-15
Catalyst CSB: Ni-Bi-O/SBA-15
Pore diameter
(nm)
Pore surface
area (m2/g)
Pore volume Pore surface
Pore volume
(cm3/g-
(cm3/g)
area (m2/g-
support)
support)
1.7-3.9
3.9-5.6
5.6-266
Total
714 (66.1)
60 (5.5)
0.523 (48.4)
0.066 (6.1)
0.491 (45.5)
1.080 (100)
213 (43.2)
53 (10.9)
225 (45.9)
491 (100)
0.096 (18.5)
0.060 (11.5)
0.364 (70.0)
0.520 (100)
306 (28.4)
1080 (100)
peaks at 2θ = 28.95°, 32.33° and 33.38° with (3,1,1), (0,2,0) and
(0,0,2) diffraction lines, respectively, were identified in Ni-Bi-O/SBA-15
catalyst. This new phase assignment as Bi2SiO5 agreed with the reports
in the literature [44–48].
To further study the degree of the Bi2SiO5 phase formation, the
metal loadings of Ni and Bi on SBA-15 were varied and the results are
presented in Fig. 3. The extent of the new phase formation depended on
the presence of Ni and mainly Bi species. For NiO/SBA-15 catalyst, only
NiO phase (JCPDS 00-1159) diffraction peaks at 2θ = 36.8° and 42.9°
were identified. Though an optimum loading for pure Bi2SiO5 phase is
the 20 wt% Ni and 30 wt% Bi on SBA-15 as presented in Fig. 2, the XRD
data of Bi2SiO5 phase shown in Fig. 3 were obtained as reasonably
changing with Ni loading amount. The peak intensities ascribed to
Bi2SiO5 and beta-Bi2O3 were similar in 30 wt% Bi solely loaded on SBA-
15. The result means that both phases were derived from the co-
operation between Bi species and SBA-15, possible without Ni co-im-
pregnation. The cooperation was caused by 1st step calcination of Bi
nitrate impregnated SBA-15 at the decomposition temperature of the
nitrate. The nitrate ion partially dissolves an array of silanol inside SBA-
3.1.2. X-ray diffraction (XRD)
XRD patterns of the mesoporous and conventional silica supported
catalysts are presented in Fig.2. The patterns were recorded from
2θ = 5–90 ° and then, 2θ = 25–35 ° was selected to carefully study the
Bi oxide phases in the catalysts. For Ni-Bi-O/SiO2 catalyst, the main
phases identified are beta-Bi2O3 and Bi2SiO5 respectively at
2θ = 27.38° and 29.0°. This new phase was previously assigned as
Bi2O3-a in our previous report [27]. However, this has been confirmed
as bismuth silicate phase (JCPDS 00-0287), resulting from the strong
interaction of small Bi oxide species and silica support. All the char-
acteristic peaks associated with this new phase have been successfully
identified in the silica sol supported catalyst and the mesoporous (SBA-
15 and silica foam) supported catalysts. A pure Bi2SiO5 phase with
Fig. 1. (a) N2 adsorption-desorption isotherm; (b) pore size distribution (Blue/unfilled: SBA-15, Red/Filled: Ni-Bi-O/SBA-15). (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article).
3

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
Fig. 2. XRD profiles for 20 wt%Ni-30 wt%Bi-O/support catalyst: micro-meso-
porous SiO2: CSS, CSG, CSB, CSF.
Fig. 4. (a) Temperature programmed reduction profile for 20 Ni-30 Bi-O/sup-
port catalyst: CSG (SiO2 gel), CSS (SiO2 sol), CSB (SBA-15) and CSF (SiO2 foam).
Fig. 4(b) TPR profile for impregnation methods in Ni-Bi-O, catalyst: CSG (SiO2
gel) and EQA-CSG (SiO2 gel).
mesopores with impregnated bismuth nitrate. The Bi2SiO5 of the shell
was increased by co-impregnated nickel nitrate. The layered core faced
to mesopore was formed as NiO/beta-Bi2O3 on the shell.
3.1.3. Temperature programmed reduction
Temperature programmed reduction (TPR) was utilized to de-
termine the level of active species reducibility (reduction temperature
and extent of reduction) which is very significant for catalysts activity
and selectivity to butadiene. H2-TPR profiles of the mesoporous silica
supported catalysts are presented in Fig. 4(a) and was compared with
conventional silica gel and sol catalysts.
SiO2 gel supported catalyst (CSG) showed a reduction peak of high
intensity at 500 °C and a small reduction peak from 625 °C to 700 °C.
This ease of reduction indicates the presence of NiO and BiOx species
with large particles and having weak metal-support interaction. The
SiO2 sol supported catalyst showed a similar pattern with SiO2 gel
catalyst even though it produced a shoulder peak with high intensity
aside from the main peak at 500 °C. Mesoporous supported catalysts
(CSB and CSF) however produced a broad peak at higher reduction
temperature indicating the formation of highly dispersed NiO on non-
reducible Bi2SiO5 species (caused by the strong interaction of Bi oxide
species with mesoporous supports). This confirmed the existence of
surface heterogeneity which is having more than one species con-
tributing to the overall reduction process [40]. The amount H2 con-
sumed by the various catalysts is presented as divided into the three
temperature reduction ranges in Table 3. The three ranges, low,
medium, and high temperatures previously reported [19] were used to
maintain continuity of discussion on the effect of support. The overall
hydrogen consumption depended on the availability of easily reducible
oxide species that are weakly bonded to the support surface. The
medium temperature region reflects the stability of the redox cycle for
Fig. 3. XRD patterns for loaded species in 20 Ni/SBA-15, 30 Bi/SBA-15, 20 Ni-
10 Bi-O/SBA-15, 14 Ni-10 Bi-O/SBA-15, and 10 Ni-10 Bi-O/SBA-15 catalysts.
15 mesopores with semi-dry conversion method. On the other hand, the
peak intensity derived from Bi2SiO5 was increased by increasing the Ni
loading from 10 to 20 wt% while that of beta-Bi2O3 decreased. The
Bi2SiO5 on SBA-15 was increased by co-impregnated Ni nitrate similar
to Bi nitrate.
The well-ordered array of silanol inside SBA-15 mesopores pre-
ferably caused the Bi2SiO5 shell formation on SBA-15. Therefore, re-
verse core-shell structure with uniform mesoporosity was successfully
synthesized with semi-dry conversion method. The shell was formed as
Bi2SiO5/SBA-15 by partially dissolving an array of silanol inside SBA-15
4

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
Table 3
4f5/2 at 164.3 eV which gave good agreement with reference data of
Bi2O3 [49], and identified as beta phase using XRD result. Inner surface
Bi species showed XPS peaks of Bi 4f7/2 at 158.4 eV and 4f5/2 at
163.8 eV having 0.5 eV decrease of binding energy from reference data
of Bi2O3, which was assigned as Bi2SiO5 phase using XRD result. Bi in
this phase is slightly reduced to lower valency like Bi2O3-a where a is
estimated around 0.6 based on decreasing percentage to Ni metal’s
value.
Amount of H2 consumed in TPR of supported 20 wt% Ni-30 wt% Bi-O catalysts.
Catalyst
Amount of H2 consumed [m mol/g] (TM [ºC])
(Support)
I (350−550 °C) II (550−650 °C)
III
Total
(650−850 °C)
CSG (SiO2 gel) 3.74
1.12
1.04
1.01
0.13
2.43
2.51
4.99
4.35
4.22
CSB (SBA-15)
CSF (Foamed
SiO2)
0.88
0.70
Surface Ni species showed XPS peaks of Ni 2p3/2 at 854.4 eV which
also gave good agreement with reference data of NiO [50], which is
identified as nano-particle NiO using XRD result. Inner surface Ni spe-
cies showed XPS peaks of Ni 2p3/2 at 853.7 eV having 0.7 eV decrease of
binding energy from reference data of NiO, which was assigned for
partially reduced NiO phase using XRD result. Ni in this phase is slightly
reduced to lower valency like NiO1-b where b is estimated around 0.4
based on decreasing percentage to Ni metal’s value. These changes
suggest a moderately strong interaction between the Ni and Bi nano-
particle species with the siliceous SBA-15 support. It is also an indica-
tion of charge transfer between Ni and Bi [33].
CSS (SiO2 sol)
3.75
1.10
0.38
5.23
dehydrogenation.
The TPR profile comparison for conventional SiO2 gel supported
catalyst (20 Ni-30 Bi/SiO2) and a catalyst formed using equilibrium
adsorption EQA (4 Ni-22 Bi-O/SiO2) as previously reported [19], is
shown in Fig. 4(b). The EQA catalyst contained a decreased amount of
NiO species, hence the main component present is Bi oxide species. Two
Bi species exist on the support surface which are Bi2O3 and Bi2SiO5 and
these are responsible for the two distinct reduction peaks at 500 °C and
700 °C, respectively. The decreased Ni amount resulted in a stronger
anchoring of Bi species on the support. This has increased the difficulty
in reducibility of the EQA catalyst compared to the conventional silica
supported catalyst.
3.1.5. High-resolution transmission electron microscopy (HRTEM)
The HRTEM image of SBA-15 supported catalyst as presented in
Fig. 7 shows the size and morphology of the highly ordered pores in an
array with long 1D channels. The HRTEM image agrees with the result
of pore distribution measurement by N2 adsorption. A lattice spacing of
0.21 nm which corresponded to that of the XRD detected NiO has al-
ready been presented previously [27]. XRD undetectable NiO is dis-
persed through Bi species phase in SBA-15. This implies that the oxides,
(i) ordered silica in mesopore of SBA-15, (ii) Bi2SiO5 (and beta-Bi2O3)
ordered on the silica of SBA-15 framework and (iii) NiO species dis-
persed on the Bi species form a reverse core-shell layered catalyst
system.
To further study the reduction profile of mesoporous support and its
interaction separately with the active metal species, TPR measurement
was carried out on NiO/SBA-15 and BiOx/SBA-15 and the result is
presented in Fig. 5. Structured SBA-15 can anchor both Ni and Bi spe-
cies strongly more than the conventional silica. The TPR profile showed
the difficulty in the reducibility of NiO and Bi oxide species on SBA-15
separately still remained after co-impregnation. It is evident that
oxygen species of NiO are strongly bound to the support surface. Si-
milarly, the Bi oxide deposition on the support led to the formation of
two species similar to the case of EQA catalyst. This also agrees with the
XRD profiles discussed earlier. The TPR profiles of the various combi-
nation of Ni and Bi oxides on SBA-15 support are shown in Fig. S1
(Supplementary information). All the profiles followed the same pattern
indicating a stable active species redox cycle.
3.2. Catalyst performance evaluation
3.2.1. Schematic conversion route of n-butane to products
The proposed conversion route from the n-butane to the main
product together with the byproducts is presented in Scheme 1. The
major pathways are the 1st dehydrogenation from n-butane to butenes
(1-butene and 2-butene), 2nd step dehydrogenation to butadiene, and
cracking (to mainly ethylene, propylene, methane, CO, and CO2). The
selectivity to either dehydrogenation or cracking pathways depend on
the strength of the acid and basic sites. The role of the active metal
oxide species in enhancing the production of the desired product is
discussed in the next section.
3.1.4. X-ray photoelectron spectroscopy
The chemical states and dispersion of the active metal oxide species
in the catalysts were examined with X-ray photoelectron spectroscopy
(XPS) using the Ar-etching method. A) XRD undetectable Ni oxide
species highly dispersed on Bi2SiO5, and B) Bi2O3 layer sandwiched
with SiO2 layers in layered crystal Bi2SiO5 were successfully measured
by XPS, because XPS is selectively effective for highly dispersed oxide.
The XPS spectra are presented in Fig. 6. The spectra showed before
(upper layer) and after (lower layer) etching spectrum of Bi and Ni
species. Surface Bi species showed XPS peaks of Bi 4f7/2 at 158.9 eV and
3.2.2. Effect of bismuth oxide loading
The effect of Bi amount loaded to the Ni-Bi-O/mesoporous silica
SBA-15 system on the catalyst performance is presented as comparison
between 30 wt% and 10 wt% loading as presented in Table 4. De-
creasing the Bi amount slightly improved both n-butane conversion
(activity) and butadiene (BD) selectivity resulting to increase in BD
yield increase. This is mainly due to an increase in 2nd step dehy-
drogenation selectivity. Comparing the Bi oxide species present in 20
Ni-30 Bi and 20 Ni-10 Bi as depicted by XRD measurements (Figs. 2 and
3), the beta-Bi2O3 phase appeared in the case of 10 wt% Bi loading.
While Bi2SiO5 species is still major in the Ni-Bi-O/mesoporous silica
SBA-15 system, the presence of beta-Bi2O3 species becomes important.
The beta-Bi2O3 species improved Ni oxide species thereby enhancing
the catalyst activity with selectivity towards 2nd step dehydrogenation
pathways.
3.2.3. Effect of nickel oxide loading
The effect of Ni amount loaded to the Ni-Bi-O/mesoporous silica
SBA-15 system on the catalyst performance is presented as comparison
from 10 wt% to 20 wt% loading as shown in Table 4. The BD yield
Fig. 5. Temperature programmed reduction profiles for catalysts using SBA-15
support: (a) 20 Ni-O, (b) 30 Bi-O and (c) 20 Ni-30 Bi-O.
5

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
Fig. 6. XPS spectra for 20 Ni- 30 Bi-O on SBA-15: [a] Bi [b] Ni, upper: surface, lower: inner surface.
Table 4
Comparison of Ni/Bi as the metal amount in the supported catalyst at 450 °C
and O2/n-C4H10 = 2.0.
Catalyst
CSB
CSB
CSB
CSB
HN-HB
HN-LB
MN-LB
LN-LB
Support
SBA-15
20/30
(2.4)
SBA-15
20/10
(7.2)
SBA-15
14/10
(5.0)
SBA-15
10/10
(3.6)
Ni/Bi [wt%]/[wt%]
(Ni/Bi atomic ratio)
n-C4H10 conversion [%]
28.9
30.0
27.7
23.5
Selectivitya [C%]
DH
75.2
15.4
12.3
47.5
24.8
19.7
5.1
71.8
14.6
7.8
78.4
15.0
10.2
53.2
21.6
17.1
4.5
83.5
17.8
11.2
54.5
16.5
12.1
4.4
2-C4H8
1-C4H8
BD
49.4
28.2
28.2
0.0
PO
OC
CO
(1-C4H8 + BD)b
BD/(1-C4H8 + BD) %c
BD yield
59.8
79.4
13.7
57.2
86.3
14.8
63.4
83.9
14.7
65.7
82.9
12.8
Fig. 7. HRTEM image of 20 Ni-30 Bi-O/SBA-15 mesoporous silica catalyst.
a
DH: dehydrogenation products, BD: butadiene, PO: partial oxidation pro-
ducts, OC: oxygenates and the cracked, and CO: carbon monoxide.
slightly increase with successive increase in Ni amount due to increase
of activity, while BD selectivity slightly decreased. Comparing the Bi
oxide species present in 10 Ni-10 Bi, 14 Ni-10 Bi and 20 Ni-10 Bi as
depicted by XRD measurements (Fig. 3), the beta-Bi2O3 phase de-
creased with increasing Ni amount while Bi2SiO5 phase increase. The
presence of beta-Bi2O3 species works important role for BD selectivity
in the Ni-Bi-O/mesoporous silica SBA-15 system, while the activity and
b
c
1-butene selectivity at 1st step dehydrogenation.
Selectivity at 2nd step dehydrogenation.
2nd step dehydrogenation selectivity depend on amount of Ni oxide
species. The beta-Bi2O3 species improved Ni oxide species thereby en-
hancing selectivity toward 2nd step dehydrogenation pathways with
Scheme 1. Schematic representation of n-butane conversion route to products.
6

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
Table 5
controlled calcination as reported in Ni-Bi-O/gamma-Al2O3 catalyst
system [16,17], Ni oxide species work in synergy with the beta-Bi2O3
species to enhance BD selectivity also in the mesoporous SiO2 support
system.
Comparison of Bi/Ni 0.42 to 0.14 as sub metal amount in Ni-Bi-O/support
catalysts at 450 °C, O2/n-C4H10 = 2.0.
Catalyst
CSG-LB
CSB-LB
CSF-LB
CSS-LB
(HB)
(HB)
(HB)
(HB)
3.3. Model of catalyst system from characterization and catalyst
performance
Support
SiO2 gel
SBA-15
SiO2 foam
SiO2 sol
Ni/Bi [wt%]/[wt%]
20/10 (20/ 20/10 (20/ 20/10 (20/ 20/10 (20/
30)
30)
30)
30)
Ni/Bi atomic ratio
7.2 (2.4)
29.3
7.2 (2.4)
30.0
7.2 (2.4)
34.1
7.2 (2.4)
35.7 (35.6)
Based on the characterization by instrumental analyses with catalyst
preparation pathway and catalyst performance in oxidative dehy-
drogenation of n-butane to butadiene, the schematic drawing of the
new mesoporous mixed oxides NiO-(beta-Bi2O3)-Bi2SiO5/SBA-15
system is shown in Scheme 2. NiO hierarchically on (beta-Bi2O3)-
Bi2SiO5 nano-particle cohabitation/SBA-15 reflects the preparation
scheme for Ni-Bi-O/SBA-15 catalyst as “reverse core-shell” structure.
n- C4H10 conversion
[%]
Selectivity*1 [C%]
DH
(17.6)
(28.9)
(29.2)
67.5 (79.1) 71.8 (75.2) 69.0 (77.6) 57.4 (78.3)
16.6 (21.7) 14.6 (15.4) 18.8 (17.2) 11.3 (18.6)
2- C4H8
1- C4H8
BD
18.0 (25.5) 7.8 (12.3)
11.4 (14.8) 11.1 (18.1)
32.9
49.4
38.8
35.0 (41.6)
(31.9)
(47.5)
(45.6)
PO
32.6 (20.9) 28.2 (24.8) 31.0 (22.4) 42.6 (21.6)
26.7 (17.5) 28.2 (19.7) 30.5 (21.0) 41.2 (20.5)
OC
4. Conclusion
CO
5.9 (3.4)
0.0 (5.1)
0.5 (1.4)
1.4 (1.1)
(1- C4H8 + BD)*2
BD/(1- C4H8+ BD) %*3
BD yield
50.9 (56.4) 57.2 (59.8) 50.2 (60.4) 46.1 (59.7)
64.6 (55.6) 86.3 (79.4) 77.3 (75.4) 76.0 (69.7)
New NiO-(beta-Bi2O3)-Bi2SiO5/SBA-15 catalyst system showed
good performance in the oxidative dehydrogenation of n-butane to
butadiene, resulting from the formation of beta-Bi2O3 and Bi2SiO5
phases on the catalyst. The change in phase between beta-Bi2O3 and
Bi2SiO5 was determined by the Ni/Bi ratio. Reverse core-shell structure
with uniform mesoporosity catalytically active for ODH of n-butene to
butadiene was successfully synthesized with semi-dry conversion
method. The shell was formed as Bi2SiO5/SBA-15 by partially dissol-
ving an array of silanol inside SBA-15 mesopores with impregnated
bismuth nitrate. The Bi2SiO5 of the shell was increased by co-im-
pregnated nickel nitrate. The layered core faced to mesopore was
formed as catalytically active NiO/beta-Bi2O3 on the shell. The degrees
of formation of Bi2SiO5 and beta-Bi2O3 phase reflected in the butadiene
selectivity through reducibility as evident from TPR and XRD studies.
The catalyst with balanced loadings of 20 wt% Ni and 10 wt% Bi ex-
hibited a clear superiority in the catalytic performance compared to the
other catalysts. The existence of beta-Bi2O3 phase enhanced BD se-
lectivity of active NiO species with suppressed cracking.
9.6 (5.6)
14.8
13.2
12.5 (14.8)
(13.7)
(13.3)
*1, *2 and *3 are same as Table 4.
suppressed cracking. The mixture-balanced 20 wt% Ni-10 wt% Bi-O/
SBA-15 catalyst effectively works as NiO-(beta-Bi2O3)-Bi2SiO5/SBA-15
system for oxidative dehydrogenation of n-butane to butadiene.
3.2.4. Effect of Bi amount loaded in Ni-Bi-O/mesoporous SiO2 system on
catalyst performance
The effect of Bi amount decreased from 30 wt% to 10 wt% on the
catalyst performance was investigated for the different silica supports in
Ni-Bi-O/support system. The results are presented in Table 5. De-
creasing Bi amount showed negative effect for BD selectivity over SiO2
foam and SiO2 sol catalyst systems. The decreased caused a small im-
provement of BD selectivity in SiO2 gel catalyst system while the ac-
tivity was largely improved until equivalent to the other systems. It was
considered that these negative or minor effects of decreased Bi amount
on BD selectivity are due to the existence of beta-Bi2O3 species in the
20 wt% Ni- 30 wt% Bi-O/support shown in Fig. 2. As 1) the importance
of beta-Bi2O3 species for BD selectivity and 2) the preparation of the
XRD undetected Ni oxide with the beta-Bi2O3 species through
CRediT authorship contribution statement
G. Tanimu: Investigation, Methodology, Writing - original draft,
Writing - review & editing. A.M. Aitani: Conceptualization, Writing -
review
& editing. S. Asaoka: Conceptualization, Methodology,
Scheme 2. Preparation scheme for Ni-Bi-O/SBA-15 catalyst: NiO on hierarchical (beta-Bi2O3)-Bi2SiO5 nano-particle cohabitation/SBA-15 as “reverse core-shell”
structure.
7

G. Tanimu, et al.
MolecularCatalysis488(2020)110893
Supervision, Writing - review & editing. H. Alasiri: Conceptualization,
[16] B.R. Jermy, B.P. Ajayi, B.A. Abussaud, S. Asaoka, S. Al-Khattaf, J. Mol. Catal. A
Chem. 400 (2015) 121–131.
Resources, Project administration.
[17] B.R. Jermy, S. Asaoka, S. Al-Khattaf, Catal. Sci. Technol. 5 (2015) 4622–4635.
[18] G. Tanimu, B.R. Jermy, S. Asaoka, S. Al-Khattaf, J. Ind. Eng. Chem. 45 (2017)
111–120.
Declaration of Competing Interest
[19] G. Tanimu, S. Asaoka, S. Al-Khattaf, Mol. Catal. 438 (2017) 245–255.
[20] J. Liang, Z. Liang, R. Zou, Y. Zhao, Adv. Mater. 29 (2017) 1–21.
[21] C.-Y. Lai, J. Thermodyn. Catal. 5 (2013) 1–3.
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
[22] R. H-Acuna, R. Nava, C.L. P-Ledesma, J. L-Romero, G. A-Nunez, B. Pawelec, E.M. R-
Munoz, Mater 6 (2013) 4139–4167.
[23] A. Ramanathan, B. Subramanian, Molecules 23 (2018) 263–276.
[24] T.I. Bhuiyan, P. Arudra, M. Akhtar, A. Aitani, R. Abudawoud, M. Al-Yami, S. Al-
Khattaf, Appl. Catal. A Gen. 467 (2013) 224–234.
Acknowledgements
[25] T.I. Bhuiyan, P. Arudra, M. Akhtar, A. Aitani, R. Abudawoud, M. Al-Yami, S. Al-
Khattaf, Can. J. Chem. Eng. 92 (2014) 1271–1282.
The authors highly appreciate King Fahd University of Petroleum &
Minerals (KFUPM). Also, the involvement of Japan Cooperation Center,
Petroleum (JCCP) in this joint research is greatly appreciated.
[26] M.A. Ibrahim, M.N. Akhtar, J. Cejka, E. Montanari, H. Balcar, M. Kubu, S. Al-
Khattaf, J. Ind. Eng. Chem. 53 (2017) 119–126.
[27] G. Tanimu, A. Palani, S. Asaoka, S. Al-Khattaf, Catal. Today 324 (2019) 97–105.
[28] D. Dumitru, R. Barjega, L. Frunza, D. Macovei, T. Hu, Y. Xie, V.I. Parvulescu,
S. Kaliaguine, J. Catal. 219 (2003) 337–351.
Appendix A. Supplementary data
[29] G. Qian, D. Ji, G. Lu, R. Zhao, Y. Qi, J. Suo, J. Catal. 232 (2005) 378–385.
[30] J.K. Reddy, K. Lalitha, V.D. Kumari, M. Subrahmanyam, Catal. Lett. 121 (2008)
131–136.
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2020.110893.
[31] H.-T. Sun, A. Hosokawa, Y. Miwa, F. Shimaoka, M. Fujii, M. Mizahata, S. Hayashi,
S. Deki, Adv. Mater. 21 (2009) 3694–3698.
[32] R. Kumar, N. Enjamuri, J.K. Pandey, D. Sen, S. Mazumder, A. Bhaumik,
B. Chowdhury, Appl. Catal. A Gen. 497 (2015) 51–57.
References
[33] C. Santra, A. Auroux, B. Chowdhury, RSC Adv. 6 (2016) 45330–45342.
[34] H. Zhang, Z. Ning, H. Liu, J. Shang, S. Han, D. Jiang, Y. Jiang, Y. Guo, RSC Adv. 7
(2017) 16602–16607.
[1] IHS, Butadiene: Chemical Economics Handbook, HIS Markit, London, 2018.
[2] J.C. Gottifredi, E.L. Sham, V. Murgia, E.M. Farfa, Appl. Catal. A Gen. 312 (2006)
134–143.
[35] A.J. Ward, A.M. Rich, A.F. Masters, T. Maschmeyer, New J. Chem. 37 (2013)
593–600.
[36] A. Palani, H.-Y. Wu, C.-C. Ting, S. Vetrivel, K. Shanmugapriya, A.S.T. Chiang, H.-
M. Kao, Microporous Mesoporous Mater. 131 (2010) 385–392.
[37] A. Palani, A. Pandurangan, J. Mol. Catal. A Chem. 226 (2005) 129–134.
[38] G. Qi, L. Fu, B.H. Choi, E.P. Giannelis, Energy Environ. Sci. 5 (2012) 7368–7375.
[39] M.A. Ali, S. Asaoka, Petrol. Sci. Tech. 27 (2009) 984–997.
[40] M.R. Quddus, M.M. Hossain, H.I. de Lasa, Catal. Today 210 (2013) 124–134.
[41] S. Adamu, M.Y. Khan, S.A. Razzak, M.M. Hossain, J. Porous Mater. 24 (2017)
1343–1352.
[3] Y. Xu, J. Lu, M. Zhang, J. Wang, J. Nat. Gas Chem. 18 (2009) 88–93.
[4] B.M. Weckhuysen, D.E. Keller, Catal. Today 78 (2003) 25–46.
[5] I. Rossetti, G.F. Mancini, P. Ghigna, M. Scavini, M. Piumetti, B. Bonell, F. Cavani,
A. Comite, J. Phys. Chem. C 116 (2012) 22386–22398.
[6] E. Santacesaria, M. Cozzolino, M. Di Serio, A.M. Venezia, R. Tesser, Appl. Catal. A
Gen. 270 (2004) 177–192.
[7] A. Dejoz, J.M. Lopez Nieto, F. Marquez, M.I. Vazquez, Appl. Catal. A Gen. 180
(1999) 83–94.
[42] P.E. Boahene, K.P. Soni, A.K. Dalai, J. Adjaye, Appl. Catal. A Gen. 402 (2011)
31–40.
[8] J.K. Lee, H. Lee, U.G. Hong, Y. Yoo, Y.-J. Cho, J. Lee, H.-S. Jang, J.C. Jung,
I.K. Song, J. Ind. Eng. Chem. 18 (2012) 1096–1101.
[43] Y. Wang, M. Noguchi, Y. Takahashi, Y. Ohtsuka, Catal. Today 68 (2001) 3–9.
[44] H. Lu, Q. Hao, T. Chen, L. Zhang, D. Chen, C. Ma, W. Yao, Y. Zhu, Appl. Catal. B:
Environ. 237 (2018) 59–67.
[9] M.A. Volpe, G.M. Tonetto, H. De Lasa, Appl. Catal. A Gen. 272 (2004) 69–78.
[10] W. Liu, S.K. Lai, H. Dai, S. Wang, H. Sun, C.T. Au, Catal. Lett. 113 (2007) 147–154.
[11] R. Bulanek, A. Kaluzova, M. Setnicka, A. Zukal, P. Cicmanec, J. Mayerova, Catal.
Today 179 (2012) 149–158.
[45] K. Hayashi, K. Yameda, M. Shima, Mater. Lett. 200 (2017) 24–26.
[46] H.W. Gao, X.F. Wang, D.N. Gao, Mater. Lett. 67 (2012) 280–282.
[47] L. Zhang, W. Wang, S. Sun, J. Xu, M. Shang, J. Ren, Appl. Catal. B: Environ. 100
(2010) 97–101.
[12] M. Setnicka, P. Cicmanec, R. Bulanek, A. Zukal, J. Pastva, Catal. Today 204 (2013)
132–139.
[13] X. Wang, G. Zhou, Z. Chen, W. Jiang, H. Zhou, Microporous Mesoporous Mater. 223
(2016) 261–267.
[48] J. Lu, X. Wang, Y. Wu, Y. Xu, Mater. Lett. 74 (2012) 200–202.
[49] https://xpssimplified.com/elements/bismuth.php Retrieved: 12th February, 2019.
[50] T. Kimura, H. Imai, X. Li, K. Sakashita, S. Asaoka, M.N. Akhtar, S.S. Al-Khattaf,
Arab. J. Sci. Eng. 39 (2014) 6617–6625.
[14] C. Wang, J.-G. Chen, T. Xing, Z.-T. Liu, Z.-W. Liu, J. Jiang, J. Lu, Ind. Eng. Chem.
Res. 54 (2015) 3602–3610.
[15] J. Rischard, C. Antinori, L. Maier, O. Deutschmann, Appl. Catal. A Gen. 511 (2016)
23–30.
8