Cumene cracking over chromium oxide zirconia: Effect of chromium(VI) oxide precursors

Article abstract of DOI:10.1016/j.apcata.2014.02.005

Two types of ZrO₂ supported with chromium nitrate (CN) and ammonium chromate (AC) for cumene catalytic cracking were studied. The physical properties of the catalysts were characterized with XRD, nitrogen physisorption, IR and TGA-SDTA. The acidic properties of the catalysts were determined by 2,6-lutidine and CO adsorbed IR spectroscopy. The XRD and nitrogen physisorption analyses confirmed higher tetragonal phase of ZrO₂ and specific surface area for Cr₂O₃-ZrO₂(AC) showing the ammonium chromate performed better interaction with Zr(OH)₄ than chromium nitrate. Besides, some of the chromium nitrate did not interact with Zr(OH)₄ and formed bulk crystalline Cr₂O₃ during the calcination. 2,6-lutidine and CO adsorbed IR spectroscopies distinguished that Cr₂O₃-ZrO₂(CN) possessed a stronger Lewis acid sites, while Cr₂O₃-ZrO₂(AC) possessed a stronger Br?nsted acid sites. In situ IR spectroscopy demonstrated that Cr₂O₃-ZrO₂(CN) showed better interaction with a molecular hydrogen than Cr₂O₃-ZrO ₂(AC) in the temperature range of 263-473 K. In addition, the bulk crystalline Cr₂O₃ interacted with a molecular hydrogen to form hydrogen bonded OH groups. At 523 K, the cumene hydrocracking indicated that the activity of Cr₂O₃-ZrO₂(CN) was about 1.7-fold higher than that of Cr₂O₃-ZrO₂(AC) with the main products of propylene and benzene. The high activity and stability of Cr₂O₃-ZrO₂(CN) were due to the better interaction with hydrogen molecules which generating high number of active protonic acid sites during the reaction. The activity of Cr₂O ₃-ZrO₂(CN) is comparable with SO₄ ^2--ZrO₂ in the temperature range of 323-573 K.

Full text of DOI:10.1016/j.apcata.2014.02.005

Applied Catalysis A: General 475 (2014) 487–496
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Cumene cracking over chromium oxide zirconia: Effect of
chromium(VI) oxide precursors
N.H.R. Annuara, A.A. Jalilb,c, S. Triwahyonoa,d,∗, N.A.A. Fatahb, L.P. Teha, C.R. Mamata
a Department of Chemistry, Faculty of Science, Universiti Teknologi Malaysia, UTM, 81310 Johor Bahru, Johor, Malaysia
b Institute of Hydrogen Economy, Universiti Teknologi Malaysia, UTM, 81310 Johor Bahru, Johor, Malaysia
c Department of Chemical Engineering, Faculty of Chemical Engineering, Universiti Teknologi Malaysia, UTM, 81310 Johor Bahru, Johor, Malaysia
d Ibnu Sina Institute for Fundamental Science Studies, Universiti Teknologi Malaysia, UTM, 81310 Johor Bahru, Johor, Malaysia
a r t i c l e i n f o
a b s t r a c t
Article history:
Two types of ZrO2 supported with chromium nitrate (CN) and ammonium chromate (AC) for cumene
catalytic cracking were studied. The physical properties of the catalysts were characterized with XRD,
nitrogen physisorption, IR and TGA–SDTA. The acidic properties of the catalysts were determined by 2,6-
lutidine and CO adsorbed IR spectroscopy. The XRD and nitrogen physisorption analyses confirmed higher
tetragonal phase of ZrO2 and specific surface area for Cr2O3–ZrO2(AC) showing the ammonium chromate
performed better interaction with Zr(OH)4 than chromium nitrate. Besides, some of the chromium nitrate
did not interact with Zr(OH)4 and formed bulk crystalline Cr2O3 during the calcination. 2,6-lutidine and
CO adsorbed IR spectroscopies distinguished that Cr2O3–ZrO2(CN) possessed a stronger Lewis acid sites,
while Cr2O3–ZrO2(AC) possessed a stronger Brønsted acid sites. In situ IR spectroscopy demonstrated
that Cr2O3–ZrO2(CN) showed better interaction with a molecular hydrogen than Cr2O3–ZrO2(AC) in the
temperature range of 263–473 K. In addition, the bulk crystalline Cr2O3 interacted with a molecular
hydrogen to form hydrogen bonded OH groups. At 523 K, the cumene hydrocracking indicated that the
activity of Cr2O3–ZrO2(CN) was about 1.7-fold higher than that of Cr2O3–ZrO2(AC) with the main products
of propylene and benzene. The high activity and stability of Cr2O3–ZrO2(CN) were due to the better
interaction with hydrogen molecules which generating high number of active protonic acid sites during
the reaction. The activity of Cr2O3–ZrO2(CN) is comparable with SO42−–ZrO2 in the temperature range of
323–573 K.
Received 5 December 2013
Received in revised form 30 January 2014
Accepted 4 February 2014
Available online 12 February 2014
Keywords:
Cr2O3
ZrO2
Chromium nitrate
Ammonium chromate
Cumene cracking
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
improved by admixtures of oxoanions of p and d elements such
as chromate, carbonate, tungstate and molybdate [7–9]. They do
The operation of a modern refinery nowadays is becoming more
complex. World-wide public concern about the earth’s environ-
ment and health considerations led into several new legislative
actions all around the world. With requirement to meet clean fuels
challenge the processing configuration has to be adapted accord-
ingly [1,2]. Focusing on petrol-fuel, several processes could be
identified to perform hydrocarbon conversion such as cracking,
alkylation and isomerization [3–5].
not form bulk solutions with ZrO2, but modify the properties of
its surface. Previously, our research group reported the properties
and acid catalytic testing of MoO3–ZrO2 [10]. We have suggested
that the doublet IR bands at 1595 and 1580 cm−1 corresponding to
the tetragonal phase of ZrO2 are responsible for the high activity
of MoO3–ZrO2 in n-heptane isomerization. Whereas, the presence
of Pt was crucial for Cr2O3–ZrO2 to enhance the activity and sta-
bility of Cr2O3–ZrO2 in cumene hydrocracking [11]. Pt facilitated
the formation of protonic acid sites at doublets of 1675 + 1660
and 1650 + 1625 cm−1 through a hydrogen spill over mechanism.
For WO3–ZrO2, the heating of pyridine preadsorbed WO3–ZrO2
in the presence of hydrogen molecules decreased the intensity
of the band at 1450 cm−1 assigned to the Lewis acid sites and
increased the intensity of the band at 1540 cm−1 revealing the for-
mation of active protonic acid sites [7]. Morterra et al. investigated
the acidic sites of SO42−–ZrO2 where the presence of tetrago-
nal phase of ZrO2 increased the activity of catalyst in n-butane
Zirconia (ZrO2) is an oxide with high melting point (2973 K),
high resistance for corrosion and low thermal conductivity mate-
rial [6]. Thermal stability and surface properties of ZrO2 can be
∗
Corresponding author at: Universiti Teknologi Malaysia, Ibnu Sina Institute for
Fundamental Science Studies, Faculty of Science, 81310 UTM Johor Bahru, Johor,
Malaysia. Tel.: +60 7 5536076; fax: +60 7 5536080.
E-mail addresses: sugeng@utm.my, sugengtw@gmail.com (S. Triwahyono).
http://dx.doi.org/10.1016/j.apcata.2014.02.005
0926-860X/© 2014 Elsevier B.V. All rights reserved.

488
N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
SO42−–Zr(OH)4 at 873 K for 3 h in air. The BET surface area of the
catalyst was 120 m2/g.
isomerization [12]. While, Soultanidis et al. has concluded that the
activity of WOx–ZrO2 was strongly affected by the nature of the
support, calcinations temperature and its surface density [13]. For
example, two types of WOx–ZrO2 were synthesized using com-
mercially available amorphous ZrOx(OH)4−2x and model crystalline
ZrO2 as support precursors. WZrOH showed a maximum activity
at the tungsten surface density of 5.2 W/nm2. In contrast, WZrO2
which was prepared with crystalline ZrO2 precursor was inactive in
the reaction. Vaudagna et al. has explored the influences of tung-
sten oxide precursors on WOx–ZrO2 and Pt/WOx–ZrO2 [14]. The
ammonium metatungstate and tungstic acid solution precursors
were used to prepare WOx–ZrO2. The important difference was
related to the migration of the precursor oxoanion into the nar-
rowest pores of zirconium hydroxide, thus leading to different pore
size distributions and amount of WO3 particles on the catalyst. In
addition, the presence of Pt was necessary to obtain the best cat-
alytic activity and stability of WOx–ZrO2. Afanasiev et al. reported
for ZrO2-supported Mo and W oxides prepared by conventional
impregnation technique, molten salt method and calcinations of
Zr(OH)4 impregnated with Mo(W) salt. They concluded that the
solids which have been prepared by calcinations of impregnated
Zr(OH)4 manifested the strongest Lewis and Brønsted acidity [15].
Although, large efforts have been undertaken to find a suitable
catalyst and an effective process for the catalytic conversion of
alkanes to more valuable hydrocarbon, fundamental study on the
influence of precursors in the properties and activity of catalysts
is still an interesting subject for developing new type of catalyst.
In this study, we have prepared Cr2O3 loaded on ZrO2 with differ-
ent Cr2O3 precursors for cumene catalytic cracking. The influences
of the precursors in the physical properties, acidity and catalytic
activity are presented and discussed. The surface analyses showed
that Cr2O3–ZrO2 prepared with ammonium chromate solution (AC)
possessed higher crystallinity, surface area and Brønsted acidity
than that prepared with chromium nitrate (CN). In contrast, the
chromium nitrate solution formed Cr2O3–ZrO2 with strong Lewis
acid sites and high ability to interact with a molecular hydrogen
in the OH stretching and Cr=O stretching regions. These proper-
ties may differ the activity of Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC)
in the cumene catalytic cracking. The effects of the presence
of bulk crystalline Cr2O3 in the activity of Cr2O3–ZrO2 are also
discussed.
2.2. Characterization
X-ray diffraction (XRD) analysis was used to determine the crys-
tallinity of the catalyst with a Bruker Advance D8 X-ray powder
´
˚
diffractometer with a Cu K␣ (ꢀ = 1.5418 A) radiation as a diffracted
monochromatic beam at 40 kV and 40 mA. The data were collected
at room temperature over the range of 2ꢁ = 2–40◦ with a scan rate
of 0.025◦ continuously. The fraction of tetragonal and monoclinic
phases of ZrO2 was determined based on Toraya equation [17].
1.31Xm
1 + 0.31Xm
Vt = 1 −
(1)
(2)
ꢀ
ꢁ
¯
Im
1
1
1
+ Im
1 1 1
( )
ꢀ
ꢁ
Xm
=
Im
¯
1
1
1
+ Im
1
1
1 + I
1 1 1
( )
t
(
)
where Xm is the intensity ratio of monoclinic ZrO2. It(1 1 1), Im(1 1 1)
ꢀ
ꢁ
¯
1
and Im
1
1
are the integrated intensity of the (1 1 1) reflection
of the tetragonal phase at 2ꢁ = 30.2◦, (1 1 1) reflection of the mono-
ꢀ
ꢁ
clinic phase at 2ꢁ = 31.8◦ and
1
1
reflection of the monoclinic
¯
1
phase of ZrO2 at 2ꢁ = 28.2◦, respectively. While 1.31 is the Toraya’s
theoretical deviation from linearity value [17].
The BET specific surface area and pore distribution of the cata-
lysts were determined by nitrogen physisorption with a Beckman
Coulter SA 3100. Approximately 0.05 g of catalyst was put into a
sample tube holder, followed by evacuation at 573 K for 3 h. The
adsorption of nitrogen then was carried out at 77 K.
Thermal analysis TGA–SDTA of the catalysts were carried out
using a Mettler-Toledo thermal analyzer, TGA/SDTA851e. The tem-
perature of the furnace was set within the range of 298–1273 K
with an increment of 5 K/min in a stream of nitrogen.
For measurement of the IR spectra, about 0.07 g catalyst was
ground and pressed in a hydraulic press (5000 psi) in order to
obtain 13 mm diameter of self-supporting wafer before placed in
the purpose-made stainless steel IR cell with CaF2 windows. The
cell is connected to a vacuum-adsorption apparatus. Prior to the
adsorption measurements, the catalyst was activated by heating
at 598 for 1 h under oxygen stream followed by heating at 598
for 3 h under hydrogen stream and outgassing at 598 K for 3 h.
The interaction of hydrogen and catalyst was observed in situ in
the temperature range of 173–473 K where 13.3 kPa of hydrogen
was introduced into the activated catalysts. While, 2,6-lutidine was
used as a basic probe molecule for evaluating of the acidity of the
catalysts, particularly in the observation of weak Brønsted acid sites
and acidic centers of Lewis acid sites. The activated catalyst was
exposed to 0.53 kPa of 2,6-lutidine at room temperature, followed
by outgassing at room temperature and 373 K. In addition, carbon
monoxide was also used as a basic probe molecule to evaluate the
acidity of the catalyst where the weak electron-donating CO may
form H-bonded complexes with the OH groups [18]. Partial pres-
sure of 13.3 kPa was introduced into the activated catalyst at room
temperature and 173 K. All spectra were recorded on an Agilent
Cary 640 FTIR Spectrometer with a spectral resolution of 4 cm−1
and with 128 scans.
2. Experimental
2.1. Catalysts preparation
Zirconium hydroxide was prepared from an aqueous solution of
ZrOCl2·8H2O (Wako Pure Chemical) by hydrolysis with 2.5% NH4OH
(Merck) aqueous solution [7]. The final pH value of the supernatant
was 9.0. The precipitate was filtered and washed with deionized
water. The gel obtained was dried at 383 K to form Zr(OH)4. The
chromium oxide loaded on zirconia (Cr2O3–ZrO2) catalyst was pre-
pared by incipient wetness impregnation technique at 353 K [16].
The aqueous chromium nitrate nanohydrate (Cr(NO3)3·9H2O) or
aqueous ammonium chromate ((NH4)2CrO4) was impregnated on
the Zr(OH)4, followed by drying overnight at 383 K and calcination
at 873 K for 3 h in air. The catalysts are denoted as Cr2O3–ZrO2(CN)
and Cr2O3–ZrO2(AC) for catalysts prepared with chromium nitrate
nanohydrate and ammonium chromate precursors, respectively.
The content of Cr2O3 was adjusted to 8 wt%. While, ZrO2 was pre-
pared by calcination of Zr(OH)4 at 873 K for 3 h in air.
2.3. Cumene cracking
Cumene catalytic cracking was carried out in a microcatalytic
pulse reactor at 323–573 K under hydrogen or nitrogen stream.
Prior to the reaction, 0.4 g portion of the catalyst was charged
into an ID10 mm tubular quartz glass reactor, and then it was
subjected to O2 treatment (O2 = 100 mL/min) at 673 K for 1 h, fol-
lowed by H2 reduction (H2 = 100 mL/min) at 673 K for 3 h. Then, the
The sulfate ion-treated Zr(OH)4, which is denoted as
SO42−–Zr(OH)4, was prepared by impregnation of the Zr(OH)4
with 1 N H2SO4 aqueous solution followed by filtration and
drying at 383 K [9]. SO42−–ZrO2 was obtained by calcination of

N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
489
Fig. 1. X-ray diffraction patterns of (A) fresh and (B) activated ZrO2 calcined at 873 K, Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC); (᭹) monoclinic phase of ZrO2; (ꢀ) tetragonal
phase of ZrO2; (*) cubic phase of ZrO2; (ꢁ) bulk crystalline Cr2O3.
reactor was cooling down to a reaction temperature under the car-
rier gas stream. A dose of reactant (14 ␮mol) was passed over the
activated catalyst and the products were trapped at 77 K before
being flash-evaporated into an online 6090N Agilent Gas Chro-
matograph equipped with VZ7 packed Column and FID detectors.
The intervals between doses were kept constant at 30 min. The
reaction reached steady state at pulse number four (120 min). The
activity of the catalyst was also evaluated in the reaction tempera-
ture range of 323–573 K.
a crystalline Cr2O3 during the calcinations [11]. While, almost
no-peak corresponds to the bulk crystalline Cr2O3 was observed
on Cr2O3–ZrO2(AC). In general, the presence of chromate anion
which interacted with Zr(OH)4 inhibited the sintering of ZrO2
crystallites and led to stabilize the tetragonal phase of ZrO2
[7,21,22]. Table 1 presents the relative volume of monoclinic
and metastable tetragonal phases of ZrO2 (M/T) for the catalysts.
ZrO2 exhibited a highest ratio of monoclinic over tetragonal
phase followed by Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC). In fact,
Cr2O3–ZrO2(AC) exhibited higher tetragonal phase at 30.2◦ than
that of Cr2O3–ZrO2(CN), indicating the ammonium chromate has
better interaction with Zr(OH)4. This may be due to the higher
hydrophilicity of ammonium chromate than chromium nitrate
nonahydrate [23]. The presence of monoclinic phase on ZrO2,
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) indicated that there existed
free Zr(OH)4 hydroxyl groups on the catalyst which sintered to
form monoclinic phase of ZrO2 during the calcination. A similar
result was reported by Trunschke et al. in which the monoclinic
and tetragonal phase of ZrO2 coexisted for 0.5–2 wt% chromium
loading on ZrO2. While, the tetragonal modification of ZrO2
predominated for 4 wt% chromium loading and above [24].
In addition to the fresh catalysts, Fig. 1B shows the XRD pat-
terns of the activated ZrO2, Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC)
at 673 K. The ratio of the monoclinic to tetragonal phase of ZrO2
was quite similar to the fresh catalysts, although the peaks became
broader and lower in intensity. The changes were probably due to
the inhomogeneous composition in a solid solution and smaller
crystallite size in the crystalline materials after the activation pro-
cess.
The selectivity to particular product (Si) and conversion of
reactant (Xreactant) were calculated according to Eqs. (3) and (4),
respectively.
Ci
ꢀ
ꢁ
Si =
× 100;
(3)
ꢂ
Ci − Cres reactant
ꢀ
ꢁ
ꢂ
Ci − Cres reactant
ꢀ
ꢁ
Sreactant
=
× 100,
(4)
ꢂ
Ci
where Ci and Cres reactant are mole number of particular compound
and residual reactant which was calculated based on the Scott
hydrocarbon calibration standard gas (Air Liquide America Spe-
cialty Gases LLC).
3. Results and discussion
3.1. Physical properties of Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC)
Fig. 1A illustrates the XRD patterns of fresh ZrO2,
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) calcined at 873 K. All
catalysts exhibited three well established polymorphs of mono-
clinic, tetragonal and cubic phase of ZrO2. The diffraction peaks
at 2ꢁ = 30.2◦ and 35.5◦ attributed to the tetragonal phase of ZrO2,
while the peaks at 2ꢁ = 24.2◦, 28.1◦, 31.4◦, 33.9◦ and 38.4◦ attributed
to the monoclinic phase of [6,19]. A small peak at 2ꢁ = 35.2◦ cor-
responds to the cubic phase of ZrO2 [20]. There is also a weak
and broad peak appeared at 2ꢁ = 24–25◦ on Cr2O3–ZrO2(CN),
assigned to bulk crystalline Cr2O3, showing the existence of
free chromium(VI) oxide on the surface of ZrO2 which forming
Nitrogen physisorption analysis showed that the BET specific
surface areas of ZrO2, Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) were
42, 150 and 193 m2/g, respectively (Table 1). This indicated that
the introduction of Cr2O3 increased the surface area of the catalyst
due to the formation of small diameter of pores in the catalysts.
Whereas, the C (constant) values of ZrO2, Cr2O3–ZrO2(CN) and
Cr2O3–ZrO2(AC) were 63.69, 55.10 and 54.86, respectively, show-
ing the catalysts are oxide or mixture of oxide materials. In Fig. 2A,
the N2 adsorption–desorption isotherm could be classified as a

490
N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
Table 1
Properties of ZrO2, Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC).
Moniclinic/tetragonal
phase of ZrO2
Gaussion deconvolution area of
2,6-lutidine adsorbed FTIR
Sample
BET surface
area [m2/g]
Total pore
C (constant)
value
volume [cm3/g]
Fresh
Reduced (at
673 K)
Brønsted acid sites
1640 + 1630 cm−1
Lewis acid sites
1567 + 1551 cm−1
ZrO2
Cr2O3–ZrO2(CN)
Cr2O3–ZrO2(AC)
42
150
193
0.162
0.188
0.172
63.69
55.10
54.86
83/17
49/51
25/75
82/18
48/52
24/76
–
–
0.7799
3.0999
2.2760
1.8400
with higher surface area (165 m2/g) compared to the ammo-
nium chromate (145 m2/g). In terms of the reactivity properties
in the dehydrogenation of propane, Cr2O3–Al2O3 prepared with
chromium nitrate exhibited higher selectivity than that of prepared
with ammonium chromate.
type IV isotherm with the pore filling restricted to a range of
P/P0 = 0.4–0.7 assigned to mesoporous materials (intraparticles).
The hysteresis loop was a H2 type indicating an agglomeration
of uniform spheres with a complex mesoporous structure [25].
For Cr2O3–ZrO2(CN), the size of the hysteresis loop was bigger
than those of ZrO2 and Cr2O3–ZrO2(AC). This may be due to the
bigger capillary condensation after adding the ammonium chro-
mate precursor on the ZrO2. The amount of adsorbed nitrogen on
Cr2O3–ZrO2 increased by about 20% at P/P0 = 0 than ZrO2 indi-
cating the introduction of Cr2O3 altered some mesoporous to a
microporous structure. The increase was extensively observed on
Cr2O3–ZrO2(AC) than Cr2O3–ZrO2(CN). Simultaneous decrease was
observed at P/P0 = 0.9, which reflects to the interparticle textural
porosity [26].
The thermal behaviors of Zr(OH)4, Cr2O3–ZrO2(CN) and
Cr2O3–ZrO2(AC) are represented in Fig. 3. Negative DTA peak
indicated an endothermic process whereas the positive peak rep-
resented an exothermic process. The endothermic peak at about
340 K was a result of sample melting during dehydration as weight
loss and water emission. The first weight loss showed by TGA curve
arose in the range between 300 and 417 K due to the emission of
water [28,29]. Another DTA peak at 465 K corresponds to the dehy-
droxylation of Zr(OH)4 produced an amorphous ZrO2. A small peak
was also observed at 670 K which may correspond to the transfor-
mation of amorphous material into crystals [30]. DTA peaks at 770
and 940 K for both Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) may be
due to the glow exotherm. It was found to be present in metal oxide
which commonly attributed with the transition of an initially amor-
phous phase into a crystalline modification of ZrO2. The exothermal
process probably corresponds to the crystallization of tetragonal
metastable phase of ZrO2 [31]. There was a small weight loss at
high temperature showing the dehydroxylation process on both
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC). In addition, the decrease of
TGA curve was observed extensively on Cr2O3–ZrO2(CN) which
may associated with the continuous dehydration of the solid [32].
Fig. 4 presents the IR spectra in the OH groups stretching region
at 3900–3000 cm−1 when Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC)
were activated at different temperatures. When the catalysts
were activated at 473 K, there were strong absorbance band at
3770 cm−1, shoulder band at 3625 cm−1 and broad band in the
Fig. 2B shows the pore size distribution of the catalysts
where ZrO2 has smaller total pore volume compared to both
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) (Table 1).
A significant
change in the mesopores was observed for Cr2O3–ZrO2(CN) and
Cr2O3–ZrO2(AC) in which the loading of Cr2O3 decreased the num-
bers of pores with diameter of 5–10 nm and markedly increased
the pore with diameter less than 5 nm. Inset figure shows that
ZrO2 demonstrated the presence of bigger pore volume in the
range of 20–60 nm related to the higher nitrogen adsorption at
P/P0 = 0.9 indicating that higher partial pressure was associated
with a smaller particle size and total pore volume [26]. Although,
the total pore volume of Cr2O3–ZrO2(AC) was slightly lower than
that of Cr2O3–ZrO2(CN), Cr2O3–ZrO2(AC) has a higher surface area
due to the higher number of micropore and external surface area
which referring to the Cr2O3 metal deposition on the pores [27].
However, different results was observed on Cr2O3–Al2O3 in which
Cherian et al. reported that the chromium nitrate formed catalyst
Fig. 2. (A) Nitrogen adsorption–desorption isotherm; solid symbol, adsorption; open symbol, desorption. (B) Pore size distribution of ZrO2, Cr2O3–ZrO2(CN) and
Cr2O3–ZrO2(AC). Inset: pore diameter of 20–100 nm for the catalysts.

N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
491
Fig. 3. DTA and TGA curves of Zr(OH)4, Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC).
range of 3600–3000 cm−1 for both catalysts. The bands centered
at 3770 and 3625 cm−1 showed the existence of several hydroxyl
stretching bands with different acidic centers and strength. These
bands were attributed to bibridged and tribridged OH groups cor-
respond to the presence of tetragonal and monoclinic phases of
ZrO2 on the catalyst [33]. The high intensity of the bibridged OH
groups on Cr2O3–ZrO2(AC) signified the presence of higher tetrago-
nal phase of ZrO2 in accordance with the XRD result. Whereas
the broad band in the stretching region of 3600–3000 cm−1 rep-
resented weak OH groups which interacted with some species or
defect structure on the surface. At high activation temperature,
this broad band disappeared, contrarily the bands centered at 3770
and 3625 cm−1 sharped due to the removal of weak OH groups.
In addition, high activation temperature developed a new band
centered at 3460 cm−1 attributed to the non-acidic OH groups for
Cr2O3–ZrO2(AC) [34]. Similar results were observed on WO3–ZrO2
[35] and MoO3–ZrO2 [10] in which increasing the activation tem-
perature eliminated or decreased the OH groups at stretching
region of 3800–3400 cm−1
.
3.2. Interaction of hydrogen
Hydrogen molecules play an important role in the promotion
effects of acid-catalyzed reaction where it can be rationalized by
the formation of active protonic acid sites through a molecular
hydrogen spillover phenomenon [36]. For example, the presence
Fig. 4. IR spectra of activated Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC). Samples were activated at 473, 523, and 573 K.

492
N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
Fig. 5. IR spectra of activated Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) in the OH stretching region when the catalysts were exposed to 13.3 kPa hydrogen at 173 K, followed by
heating at (b) 173 K, (c) 193 K, (d) 223 K, (e) 248 K, (f) 263 K, (g) 273 K, (h) 303 K (i) 323 K, (j) 373 K, (k) 423 K, and (l) 473 K. (a) The catalysts were activated at 598 K.
of molecular hydrogen enhances hydroisomerization and hydro-
cracking over SO42−–ZrO2, MoO3–ZrO2 and WO3–ZrO2 catalysts.
The molecular hydrogen is adsorbed–desorbed on the specific
active sites to form hydrogen atoms, followed by spillover onto the
surface support to form acidic OH groups in which the acidic OH
groups act as active sites in the reaction. In Figs. 5 and 6, the inter-
action of hydrogen molecules with the surface of Cr2O3–ZrO2(CN)
and Cr2O3–ZrO2(AC) was observed in the OH stretching and Cr=O
stretching regions in the temperature range of 173–473 K.
Fig. 5 shows the IR spectra of OH groups stretching region for
molecular hydrogen interacted with the surface samples in the
temperature range of 173–473 K. For all temperature range, only
broad band corresponds to hydrogen bonded OH groups inter-
acted with some species in the surface was observed for both
Fig. 6. IR spectra of (A) Cr2O3–ZrO2(CN) and (B) Cr2O3–ZrO2(AC) in the Cr=O stretching region when the catalysts were exposed to 13.3 kPa hydrogen at (b) 173 K, (c) 193 K,
(d) 223 K, (e) 248 K, (f) 263 K, (g) 273 K, (h) 303 K (i) 323 K, (j) 373 K, (k) 423 K, and (l) 473 K. (a) The catalysts were activated at 598 K. (C) and (D) show the subtracted spectra
of Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) in (A) and (B). The spectra were subtracted with the activated catalysts (spectrum (a) in (A) and (B). Dotted lines; spectra of the
subtracted activated samples.

N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
493
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC). The adsorption of molecu-
lar hydrogen at 173 K may be conferred the physical adsorption
of molecular hydrogen centered at 3200 cm−1. Almost no-change
was observed for the bibridged and tribridged OH groups. As the
temperature increased, the absorbance band centered at 3200 cm−1
eroded and new broad band centered at 3550 cm−1 evolved. At and
above 303 K, the absorbance band at 3200 cm−1 disappeared with a
concomitant intensification in the broad and wide absorbance band
centered at 3550 cm−1. The evolution of the absorbance bands may
be due to the alteration of the physical adsorption of molecular
hydrogen at low temperature to the hydrogen bonded OH groups at
relatively higher temperature [37]. The formation of the hydrogen
bonded OH groups began to occur at 263 K on both catalysts indicat-
ing the adsorption and dissociation of molecular hydrogen to form
hydrogen atoms occurred at relatively low temperature over both
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC). In addition, the absorbance
bands correspond to the bibridged and tribridged hydroxyl groups
at 3770 and 3625 cm−1 shifted to a lower wavenumbers due to the
inductive effect of hydrogen species coordinated to the cus Zr4+
sites in the vicinity of Cr2O3 [38]. A similar result was observed on
the adsorption of molecular hydrogen over PtHZSM5 at 173–573 K
in our previous assignment. The raising of the hydrogen adsorption
temperature of PtHZSM5 eliminated OH groups at 3700, 3610 and
3520 cm−1 with a concomitant development of new OH groups at
3690, 3675, 3600 and 3380 cm−1. This indicated that the adsorbed
hydrogen species exhibited different characteristics as the tem-
perature was varied. At 248 K, the adsorbed molecular hydrogen
began to dissociate into atomic hydrogen species over the PtHZSM5
[39,40].
Fig. 6 illustrates the variations of IR spectra of activated
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) in the Cr=O stretching region
in contact with molecular hydrogen at different temperatures.
Curve (a) in Fig. 6A and B shows the spectra of the activated catalysts
before hydrogen exposure in which both catalysts possessed bands
at 1030 and 1010 cm−1 showing the presence of two independent
Cr=O species. No-dioxo O=Cr=O species existed on the activated
catalysts due to the absence of symmetric and antisymmetric
stretching modes with different wavenumber of 30–50 cm−1 on
both catalysts [35]. Another weak bands was observed at 965, 950
and 910 cm−1 corresponding to the presence of bulk crystalline
Cr2O3 on Cr2O3–ZrO2(CN). No distinct band corresponds to the bulk
crystalline Cr2O3 on Cr2O3–ZrO2(AC). The changes of absorbance
bands in contact with a molecular hydrogen in the temperature
range of 173–473 K are shown in curve (b) to (l) in Fig. 6A and B. No
significant changes of the spectra were observed at and below 248 K
except small changes in the background line below 1000 cm−1 and
the crystalline Cr2O3 bands. The background slightly lifted and
the band of crystalline Cr2O3 disappeared showing the hydrogen
species interacted with the surface of catalyst and crystalline Cr2O3.
The obvious changes were observed at and above 263 K in which
the bands at 1030 and 1010 cm−1 decreased in intensity and shifted
to lower frequencies of 1025 and 1005 cm−1. Similar to those of the
bibridged and tribridged OH groups, the shifting of Cr=O stretch-
ing bands to lower frequencies are due to the inductive effect of
hydrogen species coordinated to the cus Zr4+ sites in the vicinity of
Cr2O3. In addition, a new broad and unresolved band was observed
at 995–960 cm−1 attributed to the hydrogen bonded OH groups in
the surface of catalysts. The alterations of the spectra caused by
the hydrogen interaction are more clearly seen in Fig. 6C and D in
which the spectra are subtracted with the activated catalyst.
Fig. 6C and D shows the subtracted IR spectra of Fig. 6A and B.
All spectra were subtracted with the activated sample and dotted
curve (a) in Fig. 6C and D shows the spectrum of subtracted spectra
of activated sample. H2-exposure at low temperatures (173–248 K)
for both catalysts showed a wide broad band at stretching region
of 1050–900 cm−1 signifying the presence of hydrogen species
interacted with some species in the surface. Particularly, small
absorbance band was observed at 1030 and 1010 cm−1 on both
catalysts, while small bands at 965, 950 and 910 cm−1 were only
observed on the Cr2O3–ZrO2(CN). The bands at 965, 950 and
910 cm−1 is suggested to the hydrogen species coordinated to
the bulk crystalline Cr2O3. Significant changes were observed at
and above 263 K in which the broad band at 1050–900 cm−1
eroded and the bands at 1030 and 1010 cm−1 shifted to 1035 and
1015 cm−1. In addition, the bands at 965, 950 and 910 cm−1 on
the Cr2O3–ZrO2(CN) intensified considerably. The changes may be
corresponded to the alteration of the hydrogen species (hydrogen
bonded H2–O) to the hydrogen bonded OH groups at and above
263 K [41]. Cr2O3–ZrO2(CN) showed a higher intensity of the bands
at 1035 and 1015 cm−1 at high temperature, while Cr2O3–ZrO2(AC)
showed a higher intensity at low temperature. This result suggested
that Cr2O3–ZrO2(CN) interacted with more molecular hydrogen
to form hydrogen bonded OH groups at high temperature than
Cr2O3–ZrO2(AC). In addition, the bulk crystalline Cr2O3 interacted
with hydrogen species at wide range of adsorption temperatures.
It can be suggested that molecular hydrogen dissociates on the
reduced chromium species to form hydrogen atoms, and the hydro-
gen atoms migrate by spillover and diffusion on the Cr2O3–ZrO2
surface where they are converted into protonic acid sites (H+) and
hydride (H−). H+ is formed by releasing an electron and stabilized
on atomic O near the Lewis acidic center. ZrO2 support (which Zr4+
being the Lewis site) traps an electron and then reacts with a sec-
ond spiltover hydrogen to form an H− which bonded with the Lewis
acid site [3]. It seems that the dissociation of H2 does not occur
on ZrO2 or cus Zr4+, since the interaction of H2 and ZrO2 does not
form protonic acid sites and/or hydride. In fact, the presence of bulk
crystalline Cr2O3 increased the interaction of H2 and the surface of
catalyst. Barton et al. signified that partly reduced WOx domains
on the ZrO2 may act as redox sites required for the formation of H+
species from H2. The lodging of a proton by electron transfer and
charge delocalization across an extended W–O network results in
electronic structures similar to those of heteropoly acids [42,43].
On the basis of the IR results in the present study together with
the previous results reported on WO3 loaded on ZrO2, it is sug-
gested that the absorbance band at 1030 cm−1 on the Cr2O3–ZrO2
corresponds to the stretching of the Cr=O which is connected to
cus Zr4+ through O atom [35]. While, the band at 1010 cm−1 cor-
responds to the stretching of the Cr=O which is connected to the
other Cr atom through O atom [35]. The spectra of Cr2O3–ZrO2 are
essentially the same to that of WO3–ZrO2 in our previous report in
which two independent W=O species were observed at 1021 and
1014 cm−1 after the treatment at 673 K. We concluded that the IR
bands at 1021 and 1014 cm−1 are assigned to the stretching of the
W=O which is connected to cus Zr4+ through O atom and stretch-
ing of the W=O which is connected to the other W atom through O
atom.
3.3. Intrinsic acidity of Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC)
2,6-Lutidine is known to be more sensitive probe on Brønsted
acid site than pyridine due to its higher basicity and the steric
hindrance of the methyl groups which can be used to study the rel-
atively weak Brønsted acid sites and the acidic center of Lewis acid
sites [44]. Fig. 7A and B shows 2,6-lutidine adsorbed on activated
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) at room temperature, fol-
lowed by outgassing at room temperature and 373 K. There were six
bands located at 1675, 1660, 1650, 1640, 1630 and 1625 cm−1 cor-
responding to the Brønsted acid sites and another set of six bands
were located at 1607, 1598, 1593, 1580, 1567 and 1551 cm−1 corre-
sponding to the Lewis acid sites. For the Brønsted acid region, strong
doublet bands appeared at 1640 (8a mode) and 1630 (8b mode)
cm−1 representing the protonated 2,6-lutidine species adsorbed

494
N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
Fig. 7. IR spectra of 2,6-lutidine adsorbed on Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC) at room temperature, followed by outgassing at room temperature (dotted lines) and
373 K (solid lines) at (A) Brønsted and (B) Lewis acid region. IR spectra of CO adsorbed on (C) Cr2O3–ZrO2(CN) and (D) Cr2O3–ZrO2(AC). The catalysts were activated at 598 K
and CO was adsorbed at room temperature for (c) 3 min and (d) 5 min. CO was also adsorbed on the catalysts at 173 K for (e) 5 min, (f) 10 min and (g) 15 min. Dotted lines
represents the adsorption of CO on ZrO2 at (a) room temperature and (b) 173 K. Spectrum (e) show the Gaussian deconvolution bands of the catalysts.
on Brønsted acid sites. Shoulder bands at 1675 and 1660 cm−1
were due to the 8a and 8b ring vibrational modes, whereas the
bands at 1650 (8a mode) and 1625 (8b mode) cm−1 were assigned
to the 2,6-lutidinium ions of Brønsted acid sites. For the Lewis
acid region, dual doublets were observed, indicating H-bonded 2,6-
lutidine corresponding to the monoclinic phase of ZrO2. The first
of these bands was seen at 1607 (8a mode) and 1580 cm−1 (8b
mode); the other was seen at 1593 (8a mode) and 1580 (8b mode)
cm−1. Furthermore, there were strong doublet bands correspond-
ing to the tetragonal phase of ZrO2 at 1598 (8a mode) and 1580
(8b mode) cm−1. In addition, weak shoulder bands at 1567 (8a
mode) and 1551 (8b mode) cm−1 corresponding to the tetrago-
nal phase of ZrO2 were also observed [12,45–47]. Upon heating
in vacuum at 373 K, the progressive decrease was noted in the
intensity of the bands attributed to the Lewis acid sites and only
a slight decrease for the bands attributed to the Brønsted acid sites
[45,46]. Particularly, when the catalysts were outgassed at 373 K,
Cr2O3–ZrO2(CN) exhibited stronger absorbance bands at 1567 and
1551 cm−1 compared to Cr2O3–ZrO2(AC). This is suggested to be
caused by the presence of bulk crystalline Cr2O3 or tetragonal phase
of ZrO2 accordance to the XRD analysis at 2ꢁ = 24–25 and 35.5◦.
Whereas, the band area of Brønsted acid sites at doublets of 1640
and 1630 cm−1 conferred higher value for Cr2O3–ZrO2(AC) which
may be due to the presence of large number of tetragonal phase of
ZrO2. This signified that Cr2O3–ZrO2(CN) possessed stronger Lewis
acid sites and Cr2O3–ZrO2(AC) possessed stronger Brønsted acid
sites.
et al. and Onfroy et al. reported their study of CO adsorption on
ZrO2, SO42−–ZrO2 and WO3–ZrO2 at 77 K (temperature of boiling
nitrogen), the bands at 2183, 2158 and 2138 cm−1 correspond to CO
stretching bands were observed on unmodified ZrO2 [18,45]. The
difference in the adsorption temperature of CO on ZrO2 resulted a
different result in which a low temperature of 77 K is necessary for
the CO adsorption on ZrO2.
For the Cr2O3 loaded on ZrO2 (Fig. 7C and D), no distinct
absorbance bands were observed in the CO adsorption at room
temperature for at least 5 min of CO-exposure. It is suggested
that CO did not form H-bonded complexes with the OH groups at
room temperature. So, the adsorptive interaction was studied at
low temperature in order to reveal both strong and weak charge
withdrawing center. At 173 K, the formation of a single broad and
strongly asymmetric band centered at 2150 and 2160 cm−1 for
Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC), may be due to the inter-
action of the H-bonding type. Fig. 7C(e) and D(e) show the detail
deconvolution bands of Cr2O3–ZrO2(CN) and Cr2O3–ZrO2(AC). Both
catalyst have similar absorbance bands, but differ in the band pos-
itions. The bands were Lewis acid sites of ZrO2 (LZr, 2188 cm−1),
Brønsted acid sites associated with chromate (BCr, 2170 and
2174 cm−1) and ZrO2 (BZr, 2160 and 2150 cm−1) and physisorp-
tion at 2135 and 2130 cm−1 [48]. The band at 2110 cm−1 may
correspond to the CO adsorption on cus Zr3+ [49]. The inten-
sity of LZr, 2188 cm−1 for Cr2O3–ZrO2(CN) is higher than that of
Cr2O3–ZrO2(AC) showing that Cr2O3–ZrO2(CN) exhibited stronger
Lewis acid sites. The intensity of the band did not change with
the time of CO-exposure, indicating that the Lewis acid sites at
2188 cm−1 was strong acidic groups. Whereas, the intensity of the
absorbance band corresponds to the Brønsted acid sites changed
with the time indicating the presence of medium to strong acidic
groups on both catalysts. The change was observed extensively
on Cr2O3–ZrO2(CN), showing that the Cr2O3–ZrO2(CN) possessed
weaker Brønsted acidic groups than that of Cr2O3–ZrO2(AC). In
addition to the stronger Brønsted acid sites, the Cr2O3–ZrO2(AC)
possessed higher number of acidic sites due to higher intensity of
The use of strong base molecule such as 2,6-lutidine is known to
have been suitable for probing at least the Brønsted and Lewis acidic
groups of ZrO2 based-catalyst. Besides, a small and weak electron-
donating molecule such as carbon monoxide can also be utilized
as a probe molecule to characterize the acidic groups through the
H-bonding possibility. The carbon monoxide may form H-bonded
complexes with the OH groups and be easily detected by IR spec-
troscopy following the absorption in OH and CO stretching regions.
Fig. 7C shows no-band corresponding to CO for unmodified ZrO2
at room temperature and 173 K (dotted lines). However, Morterra
the band at 2160 cm−1
.

N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
495
Fig. 8. (A) The comparison of main products for the cumene cracking at 523 K over ZrO2, Cr2O3–ZrO2(CN), Cr2O3–ZrO2(AC) and SO42−–ZrO2. (B) Products distribution of
cumene hydrocracking over Cr2O3–ZrO2(CN) at 523 K.
3.4. Cumene cracking
Cumene cracking was used to examine the activity of
Cr2O3–ZrO2 in the temperature range of 323–573 K. Previously,
we have reported cumene catalytic cracking over Pt loaded on
Cr2O3–ZrO2 in which the presence of Pt markedly enhanced the
activity and stability of Cr2O3–ZrO2 [11]. In this report, the absence
of the metal active site such as Pt is due to the elucidation of the
intrinsic properties of Cr2O3 toward cumene catalytic cracking, par-
ticularly the effect of different Cr2O3 precursors in the acid catalytic
Scheme 1. Proposed mechanism of cumene cracking over Cr2O3–ZrO2.
reaction. Fig. 8A shows the comparison of cumene cracking activ-
ity over ZrO2, Cr2O3–ZrO2(CN), Cr2O3–ZrO2(AC) and SO42−–ZrO2
at 523 K in the presence of hydrogen or nitrogen carrier gas. No
of protonic acid site generated from molecular hydrogen as illus-
product was observed for the reaction in the absence of catalyst
trated in Scheme 1. The mechanism of cumene cracking involves
regardless to the carrier gas (data not shown). Trace amount of
three following modes: (1) protonation at the ring carbon in
products were observed for the reaction in the presence of nitro-
which isopropyl group is attached, (2) beta-scission to form propyl
gen carrier gas (in the absence of hydrogen) for all catalysts. In spite
cation and benzene and (3) deprotonation of propyl cation to form
in the presence of hydrogen carrier gas, ZrO2 showed low activity
propylene and proton which will be used for the protonation at
with the products of propylene and benzene were about 0.40 and
the ring carbon (1) [51–56].
0.23 nmol/m2-cat, respectively. Whereas, Cr2O3–ZrO2(CN) exhib-
Fig. 9 shows the main products of cumene hydrocracking at
ited comparable activity to SO42−–ZrO2, but it was about 1.7-fold
523 K over Cr2O3–ZrO2(CN) as a function of reaction temperature.
higher activity than that of Cr2O3–ZrO2(AC). These results showed
The propylene and benzene products increased with increase in
the indispensability of molecular hydrogen and specific active sites
the reaction temperature indicating high temperature is necessary
for assisting in the formation of active protonic acid sites for crack-
ing process. Although, it is not certain at present, the reduced
chromium species may act as a specific active site for assisting in the
formation of active protonic active sites via hydrogen spillover phe-
nomenon. In fact, almost no cracking product was observed either
in the absence of hydrogen or chromium species indicating that
the presence of permanent Brønsted acid sites has no role in the
cumene catalytic cracking over Cr2O3–ZrO2 type catalysts. While
the presence of strong Lewis acid plays an important role in the
stabilizing of formed protonic acid sites by trapping of electrons
[50].
The product distribution of cumene hydrocracking over
Cr2O3–ZrO2(CN) is shown in Fig. 8B. The main products were
propylene and benzene with the selectivity of 48.6 and 41%,
respectively. About 10% was the by-products of toluene and ethyl-
benzene. Similar to the previous report on the cumene cracking
over SO42−–ZrO2 [3], the cracking products were composed of
benzene and propylene and no propane was formed. Although it is
not shown here, the stability of Cr2O3–ZrO2(CN) was observed for
more than 100 pulses in which the reaction reached steady state
at pulse number four. The cracking of cumene to propylene and
benzene over Cr2O3–ZrO2, is generally attributed to the interaction
Fig. 9. Effect of reaction temperature in the product distribution of cumene hydro-
cracking over Cr2O3–ZrO2(CN).

496
N.H.R. Annuar et al. / Applied Catalysis A: General 475 (2014) 487–496
for enhancing the cracking process due to the formation of more
active protonic acid sites at a higher temperature. Sohn et al. [57],
reported that the cracking reaction of n-hexane took place in strong
acid sites of CrOx–ZrO2 while the active site for the dehydrocycliza-
tion reaction of n-hexane was Cr3+. The reactivity studies reveal that
the activity of Cr2O3–ZrO2 depends upon the type of structure of
catalyst.
[11] N.H.R. Annuar, A.A. Jalil, S. Triwahyono, Z. Ramli, J. Mol. Catal. A: Chem. 377
(2013) 162–172.
[12] C. Morterra, G. Meligrana, G. Cerrato, V. Solinas, E. Rombi, M.F. Sini, Langmuir
19 (2003) 5344–5356.
[13] N. Soultanidis, W. Zhou, A.C. Psarras, A.J. Gonzalez, E.F. Iliopoulou, C.J. Kiely, I.E.
Wachs, M.S. Wong, J. Am. Chem. Soc. 132 (2010) 13462–13471.
[14] S.R. Vaudagna, R.A. Comelli, N.S. Figoli, Appl. Catal., A 164 (1997) 265–280.
[15] P. Afanasiev, C. Geantet, M. Breysse, G. Coudurier, J.C. Vedrine, J. Chem. Soc.,
Faraday Trans. 90 (1994) 193–202.
[16] M. Cherian, M.S. Rao, W-T. Yang, J-M. Jehng, A.M. Hirt, G. Deo, Appl. Catal., A
233 (2002) 21–33.
[17] H. Toraya, S.J. Yoshimura, S. Sommiya, J. Am. Ceram. Soc. 67 (1984) 119–121.
[18] C. Morterra, G. Cerrato, S. Di Ciero, Appl. Surf. Sci. 126 (1998) 107–128.
[19] L.F. Liotta, A.M. Venezia, G. Pantaleo, G. Deganello, M. Gruttadauria, R. Noto,
Catal. Today 91–92 (2004) 231–236.
[20] B.L. Kirsch, A.E. Riley, A.F. Gross, S.H. Tolbert, Langmuir 20 (2004) 11247–11254.
[21] J.R. Sohn, S.G. Ryu, Langmuir 9 (1993) 126–131.
[22] J.R. Sohn, S.G. Ryu, M.Y. Park, Y.II. Pae, J. Mater. Sci. 28 (1993) 4651–4659.
[23] X. Huang, B. Hou, J. Wang, D. Li, L. Jia, J. Chena, Y. Sun, Appl. Catal., A 408 (2011)
38–46.
[24] A. Trunschke, D.L. Hoang, J. Radnik, H. Lieske, J. Catal. 191 (2000) 456–466.
[25] A.I. Trypolskyi, T.M. Gurnyk, P.E. Strizhak, Catal. Commun. 12 (2011) 766–771.
[26] M.A.A. Aziz, A.A. Jalil, S. Triwahyono, R.R. Mukti, Y.H. Taufiq-Yap, M.R. Sazegar,
Appl. Catal., B 147 (2014) 359–368.
[27] F. Trejo, M.S. Rana, J. Ancheyta, Ind. Eng. Chem. Res. 50 (2011) 2715–2725.
[28] S. Biamino, C. Badini, J. Eur. Ceram. Soc. 24 (2004) 3021–3034.
[29] X. Turrillas, P. Barnes, D. Hausermann, J. Mater. Chem. 8 (1993) 163–168.
[30] D.A. Ward, E.I. Ko, Chem. Mater. 5 (1993) 956–969.
[31] P.D.L. Mercera, J.G. Van Ommen, E.B.M. Doesburg, A.J. Burggraaf, J.R.H. Ross,
Appl. Catal. 57 (1990) 127–148.
[32] E. Torres-Garcia, J. Mater. Res. 16 (2001) 2209–2212.
[33] K.T. Jung, A.T. Bell, J. Mol. Catal. A: Chem. 163 (2000) 27–42.
[34] L.H. Little, Infrared Spectra of Adsorbed Species, Academic Press, New York, NY,
1966.
[35] S. Triwahyono, T. Yamada, H. Hattori, Appl. Catal., A 250 (2003) 75–81.
[36] M.A.A. Aziz, N.H.N. Kamarudin, H.D. Setiabudi, H. Hamdan, A.A. Jalil, S. Triwahy-
ono, J. Nat. Gas Chem. 21 (2012) 29–36.
4. Conclusion
Cr2O3–ZrO2 was successfully prepared with two different pre-
cursors of chromium nitrate (CN) and ammonium chromate
(AC). The XRD and nitrogen physisorption analyses revealed that
Cr2O3–ZrO2(AC) possessed higher crystallinity and specific surface
area than that of Cr2O3–ZrO2(CN). This may be caused by the bet-
ter interaction Zr(OH)4 with ammonium chromate than chromium
nitrate. In fact, some of the chromium nitrate did not interact with
Zr(OH)4 and formed bulk crystalline Cr2O3 on the surface during
the calcination. The acidic properties of the catalysts confirmed
that Cr2O3–ZrO2(CN) has stronger Lewis and weaker Brønsted acid
sites compared to Cr2O3–ZrO2(AC). In addition, in situ IR spec-
troscopy revealed that Cr2O3–ZrO2(CN) interacted with molecular
hydrogen to form more hydrogen bonded OH groups than that
of Cr2O3–ZrO2(AC) at and above 263 K. Besides, the bulk crys-
talline Cr2O3 demonstrated an ability to interact with a molecular
hydrogen to form hydrogen bonded OH groups on the surface.
The higher hydrogen bonded OH groups led to exhibit the high
activity Cr2O3–ZrO2(CN) in cumene hydrocracking than that of
Cr2O3–ZrO2(AC). In addition, the activity of Cr2O3–ZrO2(CN) is
comparable with SO42−–ZrO2 acid catalyst in the temperature
range of 323–573 K.
[37] V.B. Kazansky, A.I. Serykha, B.G. Andersonb, R.A. van Santen, Catal. Lett. 88
(2003) 3–4.
[38] V.B. Kazansky, A.I. Serykh, Catal. Lett. 98 (2004) 2–3.
[39] H.D. Setiabudi, A.A. Jalil, S. Triwahyono, N.H.N. Kamarudin, R.R. Mukti, Appl.
Catal., A 417–418 (2012) 190–199.
[40] H.D. Setiabudi, A.A. Jalil, S. Triwahyono, J. Catal. 294 (2012) 128–135.
[41] H. Hattori, Stud. Surf. Sci. Catal. 138 (2001) 3–12.
Acknowledgments
[42] E. Iglesia, D.G. Barton, J.A. Biscardi, M.J.L. Gines, S.L. Soled, Catal. Today 38 (1997)
339–360.
[43] D.G. Barton, S.L. Soled, G.D. Meitzner, G.A. Fuentes, E. Iglesia, J. Catal. 181 (1999)
57–72.
[44] A. Corma, C. Rodellas, V. Fornes, J. Catal. 88 (1984) 374–381.
[45] T. Onfroy, G. Clet, M. Houalla, J. Phys. Chem. B 109 (2005) 3345–3354.
[46] F. Leydier, C. Chizallet, A. Chaumonnot, M. Digne, E. Soyer, A. Quoineaud, D.
Costa, P. Raybaud, J. Catal. 284 (2011) 215–229.
[47] V. Lebarbier, G. Clet, M. Houlla, J. Phys. Chem. B 110 (2006) 13905–13911.
[48] T.N. Vu, J. van Gestel, J.P. Gilson, C. Collet, J.P. Dath, J.C. Duchet, J. Catal. 231
(2005) 453–467.
[49] Z.Y. Ma, C. Yang, W. Wei, W.H. Li, Y.H. Sun, J. Mol. Catal. 227 (2005) 119–124.
[50] H.D. Setiabudi, A.A. Jalil, S. Triwahyono, N.H.N. Kamarudin, R. Jusoh, Chem. Eng.
J. 217 (2013) 300–309.
[51] S. Tiong Sie, Ind. Eng. Chem. Res. 31 (1992) 1881–1889.
[52] S.N. Timmiati, A.A. Jalil, S. Triwahyono, H.D. Setiabudi, N.H.R. Annuar, Appl.
Catal., A 459 (2013) 8–16.
[53] K. Ebitani, J. Konishi, H. Hattori, J. Catal. 130 (1991) 257–267.
[54] K. Ebitani, J. Tsuji, H. Hattori, H. Kita, J. Catal. 135 (1992) 607–609.
[55] J.M. Parera, N.S. Fígoli, Catalytic Naphtha Reforming, second ed., Marcel Dekker,
New York, NY, 2004.
[56] M.R. Sazegar, A.A. Jalil, S. Triwahyono, R.R. Mukti, M. Aziz, M.A.A. Aziz, H.D.
Setiabudi, N.H.N. Kamarudin, Chem. Eng. J. 240 (2014) 352–361.
[57] J.R. Sohn, S.G. Ryu, H.W. Kim, J. Mol. Catal. 135 (1998) 99–106.
This work was supported by the Universiti Teknologi Malaysia
under The Research University Grant No. 04H26 and National Sci-
ence Fellowship (N.H.R. Annuar) under the Ministry of Science,
Technology and Innovation, Malaysia. Our gratitude also goes to the
Hitachi Scholarship Foundation for the Gas Chromatograph Instru-
ments Grant.
References
[1] H. Weyda, E. Köhler, Catal. Today 81 (2003) 51–55.
[2] F. Trejo, M.S. Rana, J. Ancheyta, A. Rueda, Fuel 100 (2012) 163–172.
[3] T. Shishido, H. Hattori, J. Catal. 161 (1996) 194–197.
[4] T. Kusakari, K. Tomishige, K. Fujimoto, Appl. Catal., A 224 (2002) 219–228.
[5] K. Kubo, H. Iida, S. Namba, A. Igarashi, Catal. Commun. 29 (2012) 162–165.
[6] B.M. Reddy, V.R. Reddy, J. Mater. Sci. Lett. 19 (2000) 763–765.
[7] A.H. Karim, S. Triwahyono, A.A. Jalil, H. Hattori, Appl. Catal., A 433–434 (2012)
49–57.
[8] S. Triwahyono, A.J. Aishah, S.N. Timmiati, N.N. Ruslan, H. Hattori, Appl. Catal.,
A 372 (2010) 103–107.
[9] S. Triwahyono, Z. Abdullah, A.J. Aishah, J. Nat. Gas Chem. 15 (2006) 247–252.
[10] N.N. Ruslan, N.A. Fadzlillah, A.H. Karim, A.J. Aishah, S. Triwahyono, Appl. Catal.,
A 406 (2011) 102–112.