Epoxidation of styrene with molecular O2 over sulfated Y-ZrO2 based solid catalysts

Article abstract of DOI:10.1016/j.apcata.2010.05.038

Liquid phase epoxidation of styrene was studied over non-sulfated and sulfated yttria-zirconia (Y-ZrO₂) based solid catalysts using molecular O₂ in the presence of DMF solvent. All catalysts yielded styrene oxide and benzaldehyde as

Full text of DOI:10.1016/j.apcata.2010.05.038

Applied Catalysis A: General 383 (2010) 161–168
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Epoxidation of styrene with molecular O2 over
sulfated Y–ZrO2 based solid catalysts
Beena Tyagi∗, Basha Shaik, H.C. Bajaj
Discipline of Inorganic Materials and Catalysis, Central Salt and Marine Chemicals Research Institute,
Council of Scientific and Industrial Research (CSIR), GB Marg, Bhavnagar 364002, Gujarat, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Liquid phase epoxidation of styrene was studied over non-sulfated and sulfated yttria–zirconia (Y–ZrO2)
based solid catalysts using molecular O2 in the presence of DMF solvent. All catalysts yielded styrene
oxide and benzaldehyde as two major products. Among all the catalysts studied, sulfated Co–Y–ZrO2
catalyst resulted in the maximum (61%) styrene conversion with 80% styrene oxide selectivity. Reactions
studied in the absence of DMF solvent or in the presence of a small amount of water favored benzaldehyde
formation predominantly. Thermally regenerated catalysts showed slight lower conversion (54%) without
affecting the selectivity through three reaction cycles.
Received 11 January 2010
Received in revised form 24 May 2010
Accepted 26 May 2010
Available online 4 June 2010
Keywords:
Styrene epoxidation
Molecular O2
Y–ZrO2
© 2010 Elsevier B.V. All rights reserved.
Sulfated Co–Y–ZrO2
Solid catalysts
1. Introduction
also been reported without the need for a co-reductant [7]. How-
ever, the homogenous catalysts are difficult to separate from the
Epoxidation of olefins is a field of great interest as epoxides are
industrially important chemicals and are widely used as intermedi-
ates for several perfumery chemicals, plasticizers and sweeteners
[1]. Moreover, styrene oxide is an important organic intermediate
for fine chemicals and pharmaceuticals; it has been prepared con-
ventionallybyepoxidationofstyreneusing stoichiometric amounts
of peracids as the oxidizing agent [2]. Generally, H2O2, urea–H2O2
adducts and tert-butyl hydroperoxide (TBHP) are oxidizing agents
commonly used for epoxidation of styrene. H2O2 results in high
conversion of styrene, but poor selectivity for styrene oxide; on the
other hand TBHP [3] and urea–H2O2 adduct [4] yields high styrene
oxide selectivity (>80%), but low styrene conversion (9–18%). In
most of the cases, a sacrificial co-reductant (oxygen acceptor),
usually an aldehyde, is used to achieve reasonable yields and selec-
tivities. These oxidizing agents are corrosive and hazardous in
nature leading to either low conversion or selectivity with unde-
sirable waste. Therefore, from both environmental and economical
stand points, molecular oxygen is the most desirable oxidant for
the epoxidation of alkenes.
product and are not re-usable. An attempt of heterogenization of
chiral salen complexes on solid supports such as modified MCM-41
and SBA-15 has proved to be effective for epoxidation of styrene,
though NaOCl and PyNO were used as oxidizing agent and co-
reductant, respectively [8]. Therefore, styrene epoxidation over
environment friendly and reusable solid catalysts, using molecu-
lar O2 as oxidant is desirable. Consequently, attempts have been
made to use solid catalysts such as Ti/SiO2 [9], TS-1 [4], Ti-MCM-41
[10], Fe/SiO2 [11], Al2O3 [12], Au/MgO [13], BaO and CuO [14,15]
with H2O2, urea–H2O2 adduct and TBHP as oxidizing agents. How-
ever, studies with solid catalysts using molecular oxygen without
the need of a co-reductant are few [16–19]. The first example of aer-
obic epoxidation of styrene without a co-reductant was reported by
Tang et al. [16] using Co(OAc)2 supported on zeolite NaX, resulting
in 44% styrene conversion with 60% epoxide selectivity.
Yttria-stabilized zirconia is a well known structural ceramic that
is widely used in electrochemical devices such as solid oxide fuel
cells [20] and oxygen sensors [21]. It has also been reported as
an oxidation catalyst for partial oxidation of methane to syngas
[22–24]. Though it has been widely studied as a ceramic mate-
rial, studies of its catalytic activity are few. Zr-HMS [25], Zr-MSU-V
[26] and Zr–Mn-MCM-41 [1] have been reported for oxidation
of styrene/alkene with H2O2 and TBHP. Furthermore, modifica-
tion of zirconia with sulfate anion results in sulfated zirconia, a
potential solid acid catalyst for n-alkane isomerization at ambi-
ent temperature [27] and also for other commercially important
Cobalt ions and their complexes are well known conventional
homogeneous catalysts for the selective oxidation of alkenes using
TBHP and O2 as oxidizing agents [5,6]. Ruthenium complexes have
∗
Corresponding author. Tel.: +91 278 2567760; fax: +91 278 2566970.
E-mail addresses: btyagi@csmcri.org (B. Tyagi), hcbajaj@csmcri.org (H.C. Bajaj).
0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2010.05.038

162
B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
organic transformations such as acetylation of salicylic acid with
acetic anhydride [28], isomerization of terpenes [29] and Pechmann
reaction for coumarin synthesis [30].
For comparison with 16 mol.% Y–ZrO2, Co- and Fe(16 mol.%)
–ZrO2 have also been prepared. The resulting powders were desig-
nated as 16CoZr and 16FeZr, respectively.
In the present work, we have reported the synthesis of Y–ZrO2
by the sol–gel technique and have explored its catalytic activity for
the epoxidation of styrene using molecular O2 as an oxidant in the
presence of DMF as a solvent and in the absence of any sacrificial
reductant. Addition of transition metal co-cations namely Co and
Fe to Y–ZrO2 has been done to study the effects of these cations
on its catalytic activity. The acidity values of the Y–ZrO2 based
samples have been enhanced by sulfation using H2SO4. A detailed
systematic study including effect of sulfation and co-cations, DMF
concentration, presence of water and calcination temperature of
the catalyst has been undertaken. To the best of our knowledge
this is the first report of epoxidation of styrene using molecular
O2 in the absence of a sacrificial reductant over non-sulfated and
sulfated Y–ZrO2 based catalysts.
2.2.3. Synthesis of sulfated ZrO2 and Y–ZrO2
All catalysts prepared were sulfated by H2SO4 (1 g catalyst/15 ml
1N H2SO4) under constant magnetic stirring for 30 min. Each sul-
fated powder was oven dried at 120 ◦C for 12 h, followed by
calcination at 600 ◦C for 4 h. The resulting catalysts were designated
with S- as initial, e.g., sulfated 8 mol.% Y–ZrO2 was designated as
S-8YZr.
For comparison with sulfated Y–ZrO2, pure sulfated zirconia has
also been prepared by the two-step sol–gel technique described
previously [32]. Typically, in the first step, Zr(OH)4 was obtained
by hydrolyzing 0.025 M solution of Zr-P by the drop-wise addition
of aqueous ammonia, under constant stirring, till the pH became
10.5. The obtained gel was oven dried at 110 ◦C for 12 h and treated
with H2SO4 (1 g catalyst/15 ml 1N H2SO4) under constant stirring
for 30 min in the second step. The sulfated powder was oven dried
at 120 ◦C for 12 h followed by calcination at 600 ◦C for 4 h under
static air. The resulting white powder was designated as S-Zr.
2. Experimental
2.1. Materials
2.3. Catalyst characterization
Zirconium n-propoxide (70 wt.% in n-propanol), yttrium nitrate
hexahydrate (99.9% purity) and styrene were procured from
Sigma–Aldrich, Germany; ferric nitrate ninhydrate, n-propanol
(99% purity) and tridecane were procured from S.D. Fine Chem. Ltd.,
India; cobalt nitrate hexahydrate was from Loba Chemie Pvt. Ltd.,
India; N,N-dimethylformamide was from Qualigens Fine Chemicals
Ltd., India; and aqueous ammonia solution (25%) was purchased
from Rankem, India. Oxygen (99.9% purity) was procured from Inox
Air Product Ltd., Vadodara.
All prepared non-sulfated and sulfated catalysts after calci-
nation at 500 and 600 ◦C, respectively, were analyzed with an
X-ray powder diffractometer (Philips X’pert) using CuK␣ radiation
(ꢀ = 1.5405 Å). The samples were scanned in the 2Â range of 20–80◦
at a scanning rate of 0.4◦ s−1
.
Specific surface area, pore volume and pore size of samples
were determined from nitrogen adsorption–desorption isotherms
at 77.4 K (ASAP 2010, Micromeritics, USA). Surface area and pore
size were determined using BET equation [33]. The samples were
degassed under vacuum at 120 ◦C for 4 h, prior to measurement, to
evacuate the physisorbed moisture.
2.2. Catalyst synthesis
2.2.1. Synthesis of ZrO2 and Y–ZrO2
The bulk sulfur (wt.%) present in sulfated samples before
and after calcination was analyzed with an elemental analyzer
(Perkin–Elmer, 2400, Sr II, USA).
Diffuse reflectance spectroscopy (DRS) studies of non-sulfated
and sulfated samples were performed with a Shimadzu UV-3101PC
equipped with an integrating sphere. BaSO4 was used as the ref-
erence material. The spectra were recorded at room temperature
in the wavelength range of 200–800 nm with slit width 3 nm and
medium scan speed.
All catalysts have been prepared by the sol–gel technique. Zirco-
nium n-propoxide [Zr-P] diluted to 30 wt.% was used as zirconium
precursor in all the syntheses. ZrO2 was prepared as described in
our previous work [31]. In a typical experiment, 0.025 M solution of
Zr-P was hydrolyzed by the drop-wise addition of aqueous ammo-
nia, under constant magnetic stirring, until the pH became 10.5.
After completion of the hydrolysis, the sol was continuously stirred
for 2 h at ambient temperature for the condensation of the gel. The
gel was oven dried at 110 ◦C for 12 h, followed by calcination at
500 ◦C for 4 h under static air. The resulting white powder was
designated as ZrO2.
Y–ZrO2 catalysts were prepared by doping 8 and 16 mol.% yttria.
As 2 mol of yttrium nitrate are required for 1 mol of yttria, the molar
ratio of Y:Zr for 8 mol.% Y–ZrO2 was 0.16:0.84. Typically, an alcohol
solution of yttrium nitrate hexahydrate was added to Zr-P under
constant magnetic stirring. The solution was hydrolyzed by the
drop-wise addition of aqueous ammonia, polymerized, dried and
calcined at 500 ◦C for 4 h as described above. The resulting white
powder was designated as 8YZr. Similarly, for 16 mol.% Y–ZrO2, the
molar ratio of Y:Zr in 0.025 M solution was 0.32:0.68. The resulting
white powder was designated as 16YZr.
The scanning electron microscopic (SEM) study was done with a
scanning electron microscope (Leo seriesVP1430, UK). The samples
were coated with gold using a Polaron Sputter Coater.
2.4. Surface acidity by temperature programmed desorption
(TPD) of NH3
The total surface acidity of sulfated samples was measured by
TPD-NH3 (Micromeritics Pulse Chemisorb 2720, USA) method. The
sample (0.05 g) was taken in the reactor and activated in situ at
120 ◦C for 2 h under helium (He) gas flow. After bringing down the
reactor temperature to 40 ◦C, a mixture of 10% NH3 with He gas
was passed for 30 min. The excess physisorbed NH3 was flushed
out with pure He gas flow at 40 ◦C for 30 min. The sample was then
heated at a rate of 10 ◦C min−1 up to 800 ◦C and the volume of des-
orbed NH3 per gram was measured using a thermal conductivity
detector.
2.2.2. Synthesis of transition metal doped Y–ZrO2
Co and Fe (8 mol.%) transition metal cations have been doped in
8 mol.% Y–ZrO2 to prepare Co–Y–ZrO2 and Fe–Y–Zr, respectively.
Typically, for Co–Y–ZrO2, the molar ratio of Co:Y:Zr in 0.025 M
solution was 0.08:0.16:0.76. The procedure of the hydrolysis and
condensation of the precursors to get a polymerized gel followed
by drying and calcination at 500 ◦C for 4 h, was similar as described
above. The resulting light violet colored powder was designated as
8CoYZr. Similarly, yellow colored 8FeYZr was synthesized.
2.5. Catalytic activity: epoxidation of styrene using molecular O2
The various non-sulfated and sulfated Y–ZrO2 based catalysts
have been investigated for the liquid phase epoxidation of styrene

B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
163
using molecular O2 in the presence of DMF solvent. Typically, a
50-ml round bottom flask equipped with an efficient water con-
denser was kept in a constant temperature oil bath. Then styrene
(10 mmol), DMF (10 ml) and catalyst (100 mg, pre-activated at
450 ◦C for 4 h) were added to the flask. The reaction was initiated by
bubbling O2 at atmospheric pressure into the reaction mixture at
the rate of 7–10 ml min−1. Tridecane was used as an internal stan-
dard. The reaction mixture was magnetically stirred at 600 rpm.
After 8 h of reaction, the catalyst was separated by filtering the
reaction mixture, and the liquid organic products were quantified
using a gas chromatograph (Hewlett Packard 6890) with a flame
ionization detector, capillary column (HP-5), a programmed oven
(temperature range 50–250 ◦C) and N2 as a carrier gas. Reaction
products were also confirmed by analyzing the reaction mixture
with a gas chromatograph–mass spectrometer (GC–MS, Shimadzu
GCMS-QP-2010) having a programmed oven (temperature range
50–250 ◦C), He as a carrier gas and MS in the EI mode with a 70-eV
ion source. The reaction kinetics was monitored by withdrawing
small amounts of the reaction mixture at 1 h intervals. The conver-
sion was calculated on the basis of molar percent of styrene and
tridecane (internal standard) variation; the initial molar percent of
styrene and tridecane was divided by their GC peak area percent
to get their response factor. The unreacted moles of styrene and
tridecane were calculated by multiplying the response factor by
their GC peak area percent after the reaction. Tridecane variation
was calculated by dividing the initial molar percent of tridecane by
the molar percent of tridecane remaining. The unreacted moles of
styrene after the reaction were multiplied by tridecane variation to
get the final mol percent of styrene. The conversion and selectivity
were calculated as below:
Fig. 1. XRD patterns of non-sulfated catalysts after calcination at 500 ◦C.
let, dull grey and dark grey after calcination at 600, 700 and 800 ◦C,
respectively.
Non-sulfated samples exhibited high BET surface area
(240–475 m2/g), pore volume (0.28–0.61 cm3/g) and pore diam-
eters (51–102 Å). The surface area and pore volume decreased
and pore size increased as the pore walls collapse during thermal
treatment after calcination at 500 ◦C (Table 1). The as such sulfated
sample (S-8CoYZr) showed poor textural properties in terms of
very low surface area (2.5 m2/g) and pore volume (0.002 cm3/g).
The pore diameter of the samples was also found to decrease (from
78 to 49 Å) after sulfation. This may be attributed to the filling
of pores by sulfate species; this would make them inaccessible
for nitrogen gas adsorption, resulting in very low surface area
and pore volume. After calcination at 600 ◦C, some loss of sulfate
species occurred, resulting in a significant increase in pore size
(137 Å). Therefore, surface area and pore volume slightly increased
after calcination. The other calcined sulfated samples also showed
low surface area in the range of 2.2–3.09 m2/g.
ꢀ
ꢁ
initial mol.% − final mol.%
Styrene conversion(mol.%) =
× 100
initial mol.%
ꢀ
ꢁ
GC peak area of styrene oxide
GC peak area of all products
Styrene oxide selectivity =
× 100
The response factors of the reactant and the products were cal-
culated and were found to be constant. The response factor of
styrene was in the range of 0.9694–0.9814. The response factors
of styrene oxide and benzaldehyde calculated by the standard cali-
bration curves were 0.9681 and 0.9234, respectively, and remained
within similar ranges.
The spent catalyst was recovered from the reaction mixture by
filtration, thoroughly washed with acetone to remove the adsorbed
organic species on the surface of the catalyst and then dried at room
temperature. The dried catalyst was activated at 450 ◦C for 4 h and
was used for further reaction cycles under the similar reaction con-
ditions. The catalyst was regenerated after every reaction cycle in
a similar way.
The bulk sulfur (wt.%) present in all sulfated samples after calci-
nation at600 ◦Cwasin the rangeof14.4–19.7% (Table 1). Suchahigh
amount of sulfur content was responsible for the amorphous nature
of sulfated samples after calcination at 600 ◦C. However, calcination
at 700 ◦C resulted in drastic decline in sulfur content (3.60 wt.%) of
S-8CoYZr, which further decreased to 0.59 wt.% after calcination at
800 ◦C.
Diffuse reflectance studies clearly demonstrated the doping of
Co cations in Y–ZrO2 samples. Pure Y–ZrO2 showed a spectral band
around 240 nm in the UV region similar to pure ZrO2, whereas
8CoYZr exhibited an additional broad band at around 545 nm in
the visible region (Fig. 3a), which was attributed to octahedral
Co(II) cations bonded with the Y–ZrO2 surface. However, the sam-
ple after calcination at 500 ◦C showed the broadening of the 240 nm
band and disappearance of the 545 nm band due to the formation
of cobalt-oxide, which has also been observed by its color change
from light violet to light grey. 16CoZr catalyst also showed similar
spectral behavior before and after calcination at 500 ◦C.
DR spectra of sulfated 8CoYZr (Fig. 3b), recorded after the cal-
cination at different temperatures (200–800 ◦C), showed that the
octahedral coordination of Co(II) has changed to tetrahedral coor-
dination after sulfation followed by calcination. Initially sulfation
resulted in a blue shift (504 nm) in octahedral coordination of Co(II)
with a color change to pink due to the formation of cobalt–water
3. Results and discussion
3.1. Catalyst characterization
All non-sulfated samples were found to be crystalline in nature,
having cubic crystalline phase after calcination at 500 ◦C (Fig. 1).
The light violet color of 8CoYZr and 16CoZr samples was changed
to light and dark grey, respectively, after calcination at 500 ◦C.
All sulfated samples were found to be amorphous in nature after
calcination at 600 ◦C (Fig. 2a) due to the presence of sulfate ions. As
higher thermal energy is required for the dehydroxylation of zirco-
nia in the presence of sulfate ions, the crystallization temperature
is increased [34] and the samples were amorphous after calcina-
tion at 600 ◦C. However, calcination at higher temperature of 700
and 800 ◦C resulted in crystalline sample (S-8CoYZ) having cubic
phase (Fig. 2b). The pink color of sulfated 8CoYZr changed to vio-

164
B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
Fig. 2. (a) XRD patterns of sulfated catalysts after calcination at 600 ◦C and (b) XRD patterns of S-8CoYZr catalyst after different calcination temperatures.
Table 1
Characterization of non-sulfated and sulfated Y–ZrO2 based catalysts.
Samples
Crystallite phase
Sulfur (wt.%)
Total acidity
(mmol NH3/g)
BET surface areaa (m2/g)
Total pore volumea
(cm3/g)
Average pore diametera (Å)
8YZr
–
–
–
–
–
–
–
–
89
0.53
–
0.16
0.25
175
–
57
Cubic
8CoYZr
16YZr
16CoZr
–
107
94
135
S-Zr
S-8YZr
S-8CoYZr-600
16.6
18.2
15.9
7.24
7.77
8.21
–
2.2
3.85
–
–
40
137
Amorphous
0.002
0.011
S-8CoYZr-700
S-8CoYZr-800
3.60
0.59
2.10
0.93
–
–
–
–
–
–
Cubic
S-8FeYZr
S-16YZr
S-16CoZr
S-16FeZr
14.4
18.2
17.1
19.7
–
–
3.09
–
–
–
106
–
Amorphous
7.52
7.90
–
0.003
–
0.002
2.45
225
a
Values after calcination; non-sulfated samples after 500 ◦C; sulfated samples after 600 ◦C.
Fig. 3. (a) Diffuse reflectance spectra of ZrO2, 8YZr, 8CoYZr (as such) and 8CoYZr (500) and (b) diffuse reflectance spectra of sulfated catalyst (S-8CoYZr) before and after
calcination at different temperatures.

B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
165
attributed to the presence of oxygen vacancies; doping Zr4+ with
yttria3+ resulted in the oxygen vacancies, which favored the adsorp-
tion and activation of O2 during the oxidation reaction. However,
further increase in yttria doping (16 mol.%) did not increase the cat-
alytic activity. This hints towards the saturation in the formation of
oxygen vacancies. 16YZr showed similar styrene conversion (27%)
but lower (48%) selectivity for styrene oxide.
Sulfated samples showed significant enhancement of styrene
conversion. Pure sulfated zirconia (S-Zr) resulted in 25% styrene
conversion with 50% selectivity for styrene oxide. It is notewor-
thy that non-sulfated 8YZr also showed similar styrene conversion
(25%) with higher (65%) selectivity for styrene oxide than pure
S-Zr, whereas S-8YZr resulted in higher conversion (35%) and selec-
tivity (64%) compared to pure S-Zr catalyst. The enhancement of
catalytic activity of sulfated zirconia occurs due to the electron
withdrawing effect of sulfate species. Though all sulfated samples
were amorphous in nature, having poor surface area and pore vol-
ume, these properties did not seem to take part in the catalytic
activity for styrene oxidation under the reaction conditions studied.
However, addition of sulfate species plays a significant contribution
in the enhancement of catalytic activity of the samples. In sul-
fated Y–ZrO2, a combined effect of sulfate species and the presence
of oxygen vacancies may accelerate the oxidation reaction, thus
resulting in further enhancement of conversion and selectivity.
Table 2
Epoxidation of styrene over non-sulfated and sulfated Y–ZrO2 based catalysts using
molecular O2
a
.
Catalyst
Styrene conversion (%)
Selectivity (%)
Styrene oxide
Benzaldehyde
ZrO2
8YZr
7
57
65
68
58
48
48
57
50
64
80
50
26
76
53
58
43
35
32
42
52
52
43
50
36
20
50
74
24
47
42
25
26
27
27
19
24
25
35
61
45
34
61
43
15
8CoYZr
8FeYZr
16YZr
16CoZr
16FeZr
S-Zr
S-8YZr
S-8CoYZr
S-8FeYZr
S-16YZr
S-16CoZr
S-16FeZr
No catalyst
a
Reaction conditions: Temp. = 120 ◦C, time = 8 h, styrene = 1 g, catalyst = 0.1 g,
DMF = 10 ml, Tridecane (internal standard) = 0.1 g, O2 flow = 7 ml min−1
.
complex, [Co(H2O)6]2+, as the color change in Co(II) is influenced
by the degree of hydration. After calcination at 200 ◦C a red shift to
543 nm, for octahedral Co(II), was observed. The increase in calci-
nation temperatures from 400 to 600 ◦C resulted in the appearance
of the bands of tetrahedral (629 and 582 nm) along with octahedral
(531 nm) Co(II) ions [35] with a color change from pink to violet.
This indicated a change in octahedral coordination to tetrahedral
coordination of Co(II). However, further increase in calcination
temperature to 700 ◦C showed the absence of the tetrahedral Co(II)
band at 629 nm, only very broad bands at around 240 and 551 nm
for octahedral Co(II) were observed. A color change of the sam-
ple from violet to dull grey showed the formation of cobalt-oxide
at 700 ◦C. Further calcination at 800 ◦C resulted in the formation
of cobalt-oxide of dark grey color, which showed no bands in the
visible region.
The total surface acid site concentration of the calcined sulfated
samples measured were in the range of 7.24 to 8.21 mmol/g. Among
all the sulfated samples, S-8CoYZr showed the highest surface
acidity (8.21 mmol/g). Further calcination of the sample at higher
temperature of 700 and 800 ◦C resulted in significant decrease
in the total surface acidity (2.10 and 0.93 mmol/g, respectively)
(Table 1).
3.2.2. Effect of co-cations
8YZr having Co or Fe co-cations also showed styrene conversion
(26–27%) and selectivity (68–58%) in the similar range as observed
for 8YZr (Table 2). 16CoZr and 16FeZr samples showed lower con-
version (19–24%) and similar or higher selectivity for styrene oxide
(48–57%) as compared to 16YZr. Therefore, addition of Co and Fe
co-cation did not seem to affect the catalytic activity of Y–ZrO2
samples under the present reaction conditions. Doping with yttria
resulted in similar catalytic activity to that obtained after doping
with well known oxidizing transition metals Co and Fe co-cations.
This may be explained by the transformation of the Co(II) ions to
oxide form after calcination at 500 ◦C. The cobalt-oxide form was
not catalytically active for epoxidation of styrene, therefore the pro-
cess resulted in similar conversion and selectivity to those observed
by Y–ZrO2 catalyst.
However, sulfated co-cations doped samples showed signifi-
cant rises in both the conversion and the selectivity as compared
to sulfated Y–ZrO2 catalyst. Among the Co- and Fe-doped sul-
fated Y–ZrO2, Co-doped sulfated Y–ZrO2 showed higher conversion
(61%) and selectivity (76–80%) for styrene oxide as compared to Fe-
doped sulfated Y–ZrO2 samples, which showed lower conversion
(43–45%) and selectivity (50–53%) (Table 2). The enhancements of
conversion and selectivity in sulfated co-cations doped catalysts
were due to the combined effect of sulfation along with co-cations.
The effect of Co(II) cations as shown by DR analysis indicated that,
in sulfated Co–Y–Zr samples, tetrahedral Co(II) also plays a signifi-
cant role along with the sulfate species for the oxidation of styrene.
Furthermore, the higher activity of S-8CoYZr was also evidenced by
its highest surface acidity as measured by NH3 TPD. These results
clearly showed that the acid site concentration or surface acidity
of the catalysts is mainly responsible for the oxidation activity for
styrene. As sulfated 8CoYZr showed the highest conversion and
selectivity, this sample has been selected for further detailed study.
The microscopic study revealed a spherical surface morphol-
ogy of Y–ZrO2 (Supplementary Fig. 1). Doping with Co resulted
in agglomerization. Sulfated samples also exhibited agglomerated
spherical morphology.
3.2. Catalytic activity: epoxidation of styrene using molecular O2
The catalytic activity of various non-sulfated and sulfated
Y–ZrO2 based catalysts have been calculated for the epoxidation
of styrene using molecular O2 in the presence of DMF solvent. All
catalysts yielded only two major products; styrene oxide and ben-
zaldehyde, as confirmed by GC–MS of the reaction mixture. The
mass data showed standard fragmentation patterns correspond-
ing to styrene epoxide (m/z = 119, 91, 65, 51) and benzaldehyde
(m/z = 106, 77, 51) products in the reaction mixture.
3.2.1. Non-sulfated and sulfated Y–ZrO2 based catalysts
3.2.3. Effect of temperature and kinetic study
Table 2 shows epoxidation of styrene over non-sulfated and
sulfated catalysts. Pure ZrO2 showed lower conversion (7%) of
styrene with 57% selectivity for styrene oxide. 8YZr resulted in
sharp increases in styrene conversion to 25% and in styrene oxide
selectivity to 65%. The increased catalytic activity of Y–ZrO2 may be
The oxidation of styrene with O2 over S-8CoYZr sample clearly
revealed that styrene conversion and styrene oxide selectivity
increased from 39 to 61% and 47 to 80%, respectively, after 8 h by
increasing the temperature from 110 to 120 ◦C (Table 3). On a fur-
ther increase of the temperature to 130 ◦C, conversion remained

166
B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
Table 3
Epoxidation of styrene at different temperatures over sulfated (S-8CoYZr) catalyst
a
using molecular O2
.
Temperature (◦C)
Styrene conversion (%)
Selectivity (%)
Styrene oxide
Benzaldehyde
110
120
130
39
61
62
55
Nil
Nil
47
80
45
–
–
–
53
20
55
120b
120c
120d
∼100 + CH2Oe
–
–
a
Same as in Table 2.
Reaction in the presence of 1 ml water,.
In the presence of hydroquinone (50 mg).
In the presence of hydroquinone (50 mg) over S-8YZr catalyst.
Not detected by G.C., identified by ring test.
b
c
d
e
steady (62%), however, the selectivity for styrene oxide decreased
to 45% along with an increase in benzaldehyde formation. As the
maximum conversion (61%) and selectivity (80%) were obtained
at 120 ◦C, it was chosen as the optimized temperature for further
study.
Fig. 5. Effect of concentration of DMF solvent over S-8CoYZr catalyst on epoxidation
of styrene with molecular O2.
3.2.4. Effect of catalyst and solvent concentration
The kinetic study with respect to time showed the increase in
styrene conversion and styrene oxide selectivity with increasing
the time until 7–8 h (Fig. 4). The styrene conversion increased from
25 to 60–61% and the styrene oxide selectivity from 60 to 79–80%
from 1 to 7–8 h. Further increase in time did not affect the conver-
sion until 12 h, however, the selectivity for styrene oxide decreased
successively (65%) along with an increase in the selectivity for ben-
zaldehyde with increasing the time to 12 h. It is noteworthy that
in the present study a decrease in styrene oxide selectivity leads
to an increase in benzaldehyde along with formaldehyde forma-
tion (identification of formaldehyde is discussed later) indicating
that styrene oxide undergoes a ring-opening reaction forming
benzaldehyde along with formaldehyde with time. Besides the
reaction parameters, the selectivity of the products is strongly
influenced by the catalyst and oxidant used. For example, Meng
et al. [36] obtained benzaldehyde as the main product using O2,
whereas styrene oxide was obtained as the main product when
H2O2 was used as oxidant for epoxidation of styrene with copper
hydroxyphosphate catalyst. As styrene oxide selectivity was found
maximum after 7–8 h, this time has been chosen as the optimized
time for further study.
The maximum conversion (61%) of styrene and styrene oxide
selectivity (80%) was obtained with substrate to catalyst weight
ratio of 10:1. The reaction was also carried out in the absence of the
catalyst, this showed 15% conversion and 58% selectivity for styrene
oxide (Table 2). This indicated the auto-oxidation of styrene under
the present reaction conditions. It is interesting that pure ZrO2 cat-
alyst acted as an inhibitor in the reaction and decreased the styrene
conversion as compared to that obtained in the absence of the cat-
alyst, whereas the reaction with Y–ZrO2 catalyst resulted in the
increase, both in the conversion and styrene oxide selectivity due
to the availability of oxygen vacancies that accelerate the oxida-
tion reaction. Similar inhibiting results were observed by Opre et
al. [37] for hydroxyapatite [HAp] catalyst during the epoxidation of
styrene using molecular O2 whereas better results were obtained
with Co–HAp catalyst.
It is well known that the solvent plays a crucial role in the epox-
idation reactions; these are reported [16,18,37] to perform well in
the presence of alkyl amide solvents with molecular O2. Among the
alkyl amide solvents, DMF is reported to yield the higher epoxide
[37].
We have studied the reaction in the absence and in the presence
of selected concentrations of DMF (Fig. 5). The low concentration
of DMF (5 ml) resulted in lower conversion (54%) and selectivity
(64%). On increasing the DMF concentration to 10 ml, we found that
both conversion and selectivity increased to 61 and 80%, respec-
tively. However, on further increase in DMF concentration to 20 ml,
styrene conversion was decreased significantly to 27% with a slight
decrease in selectivity (77%) and the value remained steady with a
further increase in DMF concentration to 30 ml. It may be explained
that proper dispersion of a solid catalyst was not possible in low sol-
vent concentrations leading to lower activity. On the other hand,
an increased concentration of the solvent leads to less availability
of substrate molecules to the active sites of the solid catalyst. In the
present study, 10 ml of DMF solvent was found to be sufficient for
proper dispersion of a solid catalyst, yielding maximum conversion
and selectivity.
The reaction carried out in the absence of DMF solvent was
observed to show a slightly lower conversion of styrene (57%),
however, with predominant formation of benzaldehyde along with
formaldehyde. We have checked the formation of gaseous products
like CO2 by collecting the outlet gas in limewater and also by GC
analysis, but nothing could be detected even by doing the reaction
with a high amount of styrene (4 g). No formation of formaldehyde
Fig. 4. Kinetic study over S-8CoYZr catalyst with respect to time for epoxidation of
styrene with molecular O2.

B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
167
Scheme 1. Epoxidation of styrene via two different routes in the presence and
absence of DMF solvent and in the presence of water over S-8CoYZr catalyst using
molecular O2.
could be detected by GC or GC–MS analysis. Therefore, to confirm
the presence of formaldehyde, we performed a resorcinol ring test,
wherein the addition of aqueous (1%) resorcinol and 5–6 drops of
conc. H2SO4 to the reaction mixture formed a reddish brown ring
along with the white flocculent precipitate underneath, which dif-
fused upward. The formation of reddish brown ring along with
the white flocculent precipitate confirmed formaldehyde in the
presence of benzaldehyde, otherwise benzaldehyde forms a yellow
ring and formaldehyde forms a violet one. Formation of benzalde-
hyde along with formaldehyde has also been observed by others
[17,38,39]. Hulea and Dumitriu [38] have observed the predom-
inant formation of benzaldehyde along with formaldehyde over
mesoporous Ti-MCM-41 and Ti-beta, which may be formed either
by styrene oxide or by direct oxidative cleavage of the styrene side
chain double bond by H2O2. Ghosh et al. [39] have also observed the
higher selectivity of benzaldehyde over OMS-2 (manganese oxide
octahedral molecular sieves) with TBHP and have shown the pos-
sibility of formaldehyde as the other product, which could not be
observed in GC–MS analyses.
The results clearly exhibited the role of DMF in the formation
of styrene epoxide during the epoxidation reaction of styrene. The
reaction proceeds through two different routes in the presence and
absence of DMF over sulfated CoY–ZrO2 based catalyst (Scheme 1).
It is reported that DMF associates with tetrahedral coordinated
Co(II) sites and activates the O2 [17]. Therefore, interaction of DMF
with the catalytic sites and styrene molecule favored the forma-
tion of styrene oxide, whereas in the absence of DMF formation of
benzaldehyde occurred through the second route.
Fig. 6. Effect of calcination temperature of catalyst (S-8CoYZr) on epoxidation of
styrene with molecular O2.
Table 4
Epoxidation of styrene over regenerated sulfated (S-8CoYZr) catalyst using molec-
a
ular O2
.
Reaction cycle
Styrene conversion (%)
Selectivity (%)
Styrene oxide
Benzaldehyde
Ib
61
53
54
54
80
81
81
81
20
19
19
19
IIc
IIIc
IVc
a
Same as in Table 2 except time = 7 h instead of 8 h.
Reaction with fresh catalyst.
Reaction with regenerated catalyst.
b
c
(80%). However, further increase in calcination temperature to 700
and 800 ◦C resulted in the successive decreased in both conversion
(61–40%) and selectivity (80–68%).
The surface acidity, i.e., acid sites concentration, is important for
the catalytic activity, which in turn significantly depends on the sul-
fur content of the samples. Therefore, a loss in sulfur content with
increasing calcination temperature resulted in a significant loss in
the surface acidity, as evident from Table 1, thus in the catalytic
activity of the samples. Furthermore, DR studies of S-8CoYZr also
showed the absence of tetrahedral Co(II) in the sample calcined at
higher temperature of 700 and 800 ◦C, thus resulting in the decrease
in styrene conversion and selectivity.
3.2.5. Effect of water and calcination temperature
The reaction carried out in the presence of water (1 ml) did not
affect the styrene conversion significantly (55%) but the predom-
inant formation of benzaldehyde along with formaldehyde was
observed (Table 3). This result was in contrast with the studies
reported by others [17,40]. The effect of water molecules present
in the reaction medium was found to enhance the catalytic effect of
Co-exchanged zeolite-X [17] and hydrotalcite [40]. The DMF–water
mixture in zeolite-X has resulted in the relocation of Co(II) ions so
that more ions were available to substrate molecule at the accessi-
ble sites of the catalysts [17]. In Co-exchanged hydrotalcite [40] the
catalytic activity was due to improved dispersion of Co-complex
in the hydrotalcite host in the presence of DMF–water mixture.
However, in the present study selective formation of benzaldehyde
indicated that oxidation of styrene with O2 over sulfated CoY–ZrO2
catalyst occurred through a different route in the presence of water
molecules (Scheme 1). Water molecules may occupy the catalytic
sites and thus DMF molecules may not get the chance to interact
withthe catalyticsites and thereforetheformation ofbenzaldehyde
occurred through the second route.
The temperature, at which the catalyst is calcined, before under-
going any reaction, significantly influences its acidity. Fig. 6 shows
the effect of calcination of the catalyst at different temperatures.
The catalyst calcined at 400 ◦C for 4 h showed 54% styrene conver-
sion and 62% styrene oxide selectivity. Increases in conversion and
selectivity were observed with an increase in calcination temper-
ature from 400 to 600 ◦C. The catalyst calcined at 600 ◦C showed
maximum styrene conversion (61%) and styrene oxide selectivity
3.2.6. Catalyst regeneration
The spent catalyst recovered from the reaction mixture, washed,
dried and activated at 450 ◦C was studied for further reaction cycles
under the similar reaction conditions. The regenerated catalyst
showed a slightly lower conversion (54%), however with similar
styrene oxide selectivity (81%) as compared to the fresh catalyst.
The conversion remained steady through the next three reaction
cycles without affecting the styrene oxide selectivity (Table 4). The
DR spectrum of used catalyst was recorded after each reaction cycle
and was compared with that of fresh catalyst. The spectra presented
the similar spectral bands of Co(II) ions in UV and visible region to
those of the fresh catalyst (Supplementary Fig. 2).
4. Reaction mechanism
The mechanism of styrene epoxidation by O2 over sulfated
Y–ZrO2 based catalysts is not completely understood; however, it
seems that in pure Y–ZrO2 catalysts the oxidation occurs due to

168
B. Tyagi et al. / Applied Catalysis A: General 383 (2010) 161–168
the oxygen vacancies because of the doping of Y3+ to Zr4+ lattice.
Co-complexes are well known to bind and activate oxygen forming
Co(III)–(O2−) species. On the other hand sulfation enhances the acid
strength of Zr4+ ions and activates the O2. DMF also associates with
tetrahedrally coordinated Co(II) sites and activates the O2 [17]. The
DMF adduct, formed by the interaction of DMF with the catalytic
sites, may give rise to epoxide formation, whereas, in the absence of
DMF, no adduct is formed and the predominant formation of ben-
zaldehyde along with formaldehyde occurs through second route.
In the presence of water, water molecules cover the catalytic sites
and thus the interaction of DMF molecules with the catalytic sites
did not occur, thus resulting in the formation of benzaldehyde.
Calcination at higher temperature decreases the sulfur content of
the catalyst and also deactivates the tetrahedral Co(II) species thus
decreasing the activity of the catalyst. No conversion of styrene
was found by the addition of free-radical scavenger hydroquinone
over S-YZr and S-CoYZr (Table 3), which confirms the formation of
free-radical active oxygen species.
References
[1] M. Selvaraj, S.W. Song, S. Kawi, Micropor. Mesopor. Mater. 110 (2008) 472–
479.
[2] D. Swern, in: D. Swern (Ed.), Organic Peroxides, vol. 2, Wiley Interscience, New
York, 1971.
[3] R.V. Grieken, J.L. Sotelo, C. Martos, J.L.G. Fierro, M.L. Granados, R. Mariscal, Catal.
Today 61 (2000) 49–54.
[4] S.C. Laha, R. Kumar, J. Catal. 204 (2001) 64–70.
[5] D.E. Hamilton, R.S. Drago, A. Zombeck, J. Am. Chem. Soc. 109 (1987) 374–
379.
[6] B. Rhodes, S. Rowling, P. Tidswell, S. Woodward, S.M. Brown, J. Mol. Catal. A
Chem. 116 (1997) 375–384.
[7] R. Neumann, M. Dahan, Nature 388 (1997) 353–355.
[8] R.I. Kureshy, I. Ahmad, N.H. Khan, S.H.R. Abdi, K. Pathak, R.V. Jasra, J. Catal. 238
(2006) 134–141.
[9] Q. Yang, S. Wang, J. Lu, G. Xiong, Z. Feng, Q. Xin, C. Li, Appl. Catal. A: Gen. 194
(2000) 507–514.
[10] W. Zhang, M. Froba, J. Wang, P. Tanev, J. Wong, T.J. Pinnavaia, J. Am. Chem. Soc.
118 (1996) 9164–9171.
[11] Q. Yang, C. Li, J.L. Wang, P. Ying, X. Xin, W. Shi, Stud. Surf. Sci. Catal. 130 (2000)
221–226.
[12] N.S. Patil, R. Jha, B.S. Uphade, S.K. Bhargava, V.R. Choudhary, J. Catal. 227 (2005)
217–222.
[13] N.S. Patil, B.S. Uphade, P. Jana, S.K. Bhargava, V.R. Choudhary, J. Catal. 223 (2004)
236–239.
5. Conclusions
[14] V.R. Choudhary, R. Jha, N.K. Chaudhari, P. Jana, Catal. Commun. 8 (2007)
1556–1560.
[15] V.R. Choudhary, R. Jha, P. Jana, Green Chem. 8 (2006) 689–690.
[16] Q.H. Tang, Y. Wang, J. Liang, P. Wang, Q.H. Zhang, H.L. Wan, Chem. Commun.
(2004) 440–441.
[17] J. Sebastin, K.M. Jinka, R.V. Jasra, J. Catal. 244 (2006) 208–218.
[18] J. Liang, Q. Zhang, H. Wu, G. Meng, Q. Tang, Y. Wang, Catal. Commun. 5 (2004)
665–669.
[19] J. Jiang, R. Li, H. Wang, Y. Zheng, H. Chen, J. Ma, Catal. Lett. 120 (2008) 221–
228.
[20] P. Fedtke, M. Wienecke, M.C. Bunescu, T. Barfels, K. Deistung, M. Pietrzak, J.
Solid State Electrochem. 8 (2004) 626–632.
[21] S. Yu, Q. Wu, T.-A. Massood, C.-C. Liu, Sens. Acutators B 85 (2002) 212–218.
[22] J. Zhu, J.G.V. Ommen, L. Lefferts, J. Catal. 225 (2004) 388–397.
[23] J. Zhu, J.G.V. Ommen, H.J.M. Bouwmeester, L. Lefferts, J. Catal. 233 (2005)
434–441.
[24] J. Zhu, J.G.V. Ommen, L. Lefferts, Catal. Today 112 (2006) 82–85.
[25] A. Tuel, S. Gontier, R. Teissier, Chem. Commun. (1996) 651–652.
[26] J.M. Liu, S.J. Liao, G.D. Jiang, X.L. Zhang, L. Petrik, Micropor. Mesopor. Mater. 95
(2006) 306–311.
[27] M. Hino, S. Kobayashi, K. Arata, J. Am. Chem. Soc. 101 (1979) 6439.
[28] B. Tyagi, M.K. Mishra, R.V. Jasra, J. Mol. Catal. A: Chem. 317 (2010) 41–46.
[29] (a) B. Tyagi, M.K. Mishra, R.V. Jasra, Catal. Commun. 7 (2006) 52;
(b) B. Tyagi, M.K. Mishra, R.V. Jasra, J. Mol. Catal. A: Chem. 301 (2009) 67.
[30] (a) B. Tyagi, M.K. Mishra, R.V. Jasra, J. Mol. Catal. A: Chem. 276 (2007) 47;
(b) B. Tyagi, M.K. Mishra, R.V. Jasra, J. Mol. Catal. A: Chem. 286 (2008) 41.
[31] B. Tyagi, K. Sidhpuria, B. Shaik, R.V. Jasra, Ind. Eng. Chem. Res. 45 (2006)
8643–8650.
A number of non-sulfated and sulfated Y–ZrO2 based catalysts
including Co- and Fe-doped Y–ZrO2 catalysts were synthesized by
the sol–gel technique and were studied for liquid phase epoxida-
tion of styrene using molecular O2 in the presence of DMF solvent
and in the absence of any sacrificial reductant. All catalysts showed
styrene oxide and benzaldehyde as two major products. The reac-
tion studied in the absence of DMF solvent or in the presence of a
small amount of water strongly affects the styrene oxide selectiv-
ity, resulting in benzaldehyde formation predominantly. Sulfated
Co–Y–ZrO2 showed higher styrene conversion (61%) having 80%
styrene oxide selectivity due to the combined effect of sulfate
species along with tetrahedral Co(II). The addition of sulfate species
to enhance the surface acidity and thus the catalytic activity of
Y–ZrO2 based catalysts towards styrene oxidation plays a signifi-
cant role compared to other structural and textural properties such
as amorphous or crystalline nature, surface area and pore volume
of the catalysts under the reaction conditions studied. Thermally
regenerated catalysts showed slight lower conversion (54%) with-
out affecting the selectivity through three reaction cycles.
Acknowledgements
[32] M.K. Mishra, B. Tyagi, R.V. Jasra, J. Mol. Cat. A: Chem. 223 (2004) 61–65.
[33] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, 2nd ed., Academic
Press, New York, 1982.
[34] M.K. Mishra, B. Tyagi, R.V. Jasra, Ind. Eng. Chem. Res. 42 (2003) 5727–5736.
[35] M.G. Ferreira da Silva, J. Mater. Sci. 36 (2001) 3247–3251.
[36] X. Meng, Z. Sun, R. Wang, S. Lin, J. Sun, M. Yang, K. Lin, D. Jiang, F.-S. Xiao, Catal.
Lett. 76 (2001) 105–109.
[37] Z. Opre, T. Mallat, A. Baiker, J. Catal. 245 (2007) 482–486.
[38] V. Hulea, E. Dumitriu, Appl. Catal. A: Gen. 277 (2004) 99–106.
[39] R. Ghosh, X. Shen, J.C. Villegas, Y. Ding, K. Malinger, S.L. Suib, J. Phys. Chem. B
110 (2006) 7592–7599.
The authors are thankful to CSIR Network Programme on Catal-
ysis. We are also thankful to Analytical Science Discipline for
providing instrumental analysis, namely Dr. (Mrs.) P.A. Bhatt for
X-ray, Mr. Viral for elemental sulfur analyses and Mr. Chandrakant
C.K. for SEM analyses.
Appendix A. Supplementary data
[40] V. Rives, M.A. Ulibarri, Coord. Chem. Rev. 181 (1999) 61–120.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.apcata.2010.05.038.