A green chemistry approach to styrene from ethylbenzene and air on MnxTi1-xO2 catalyst

Article abstract of DOI:10.1039/c4ra11814f

Styrene (ST) is an industrially important commodity chemical, and design of a suitable catalyst, which provides high ethyl benzene (EB) conversion and styrene selectivity at lower temperature with sustainable activity, is one of the major challenges in th

Full text of DOI:10.1039/c4ra11814f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
A green chemistry approach to styrene from
ethylbenzene and air on MnxTi1ꢀxO2 catalyst†
Cite this: RSC Adv., 2014, 4, 57087
a
a
a
*
Sanjay Singh Negi, Aswathy Thareparambil Venugopalan, Thirumalaiswamy Raja,
a
ab
*
Anand Pal Singh and Chinnakonda S. Gopinath
Styrene (ST) is an industrially important commodity chemical, and design of a suitable catalyst, which
provides high ethyl benzene (EB) conversion and styrene selectivity at lower temperature with
sustainable activity, is one of the major challenges in the field of heterogeneous catalysis. Manganese
incorporated in titania (MnxTi1ꢀxO2) anatase lattice, prepared via the solution combustion method, was
evaluated for oxidative dehydrogenation (ODH) of EB with O2 or air. MnxTi1ꢀxO2 catalysts were
characterized by different physiochemical methods. Up to 15% Mn could be introduced into the TiO2
lattice. TEM and XRD indicate disordered mesoporosity, further confirmed by adsorption isotherm
analysis. MnxTi1ꢀxO2 catalysts were evaluated for ST synthesis from EB using air or oxygen as oxidant
between 440 and 570 ꢁC. Reaction conditions have been varied systematically, such as catalyst
composition, and EB/air/O2 flow. MnxTi1ꢀxO2 shows sustainable 55% styrene yield for 45 h without
deactivation under optimum conditions. A thorough analysis of spent catalysts demonstrates the
conversion of initial anatase phase MnxTi1ꢀxO2 to Mn3O4 supported on the rutile (R) phase of TiO2. The
above change occurs in the first few hours of reaction and the Mn3O4 on R-TiO2 phase is the active
phase of the catalyst and responsible for sustainable activity for longer duration.
Received 5th October 2014
Accepted 20th October 2014
DOI: 10.1039/c4ra11814f
www.rsc.org/advances
multiwalled carbon nanotube (MWCNT) in range of 350–450 ꢁC
at atmospheric pressure, O2/EB molar ratio varied from 1 : 1 to
Introduction
3 : 1 with best EB conversion and ST selectivity values of 80%
and 92%; however, above high activity occurs at a very low ow
rate of 2.3 mL vol% in 55 mL minꢀ1 of EB + O2 + N2.9 Sekine et al.
used pervoskite and reactions were conducted at 510 or 540 ꢁC
at atmospheric pressure in the presence of steam; molar ratio of
steam to EB was 2 or 12 with 22% ST yield for 30 min.10 Ven-
Styrene (ST) is one of the important products in the petro-
chemical and polymer industries and is a precursor to several
resins, plastics, rubbers and other copolymers.1 Since 1940, the
industrial production of ST has been performed through iron
oxide promoted by potassium catalysts, by the dehydrogenation
of ethylbenzene (EB) with steam, at 700 C.2,3 This dehydroge-
ꢁ
ꢁ
nation mechanism4,5 is endothermic (DH ¼ 124.9 kJ molꢀ1) in
nature and hence it requires high reaction temperature.
Although hydrogen is available as a side product, steam-based
processes utilize a large amount of latent heat and they pose
thermodynamic limitations.6 In addition, coke formation on
the catalyst leads to severe catalyst deactivation, and EB
conversion of less than 16% was achieved per pass.7
Different materials have been used for ST synthesis, such as,
metal oxides, carbon. Makkee et al. used Al2O3 calcined between
500 and 1200 ꢁC, and achieved 42% EB conversion and 87% ST
selectivity at 475 ꢁC for 62 h.8 Qui et al. employed ozonated
ugopal et al. has shown ceria containing hydrotalcite at 450 C
at atmospheric pressure a marginally decreasing styrene yield
from 47 to 45% over a period of 72 h.11 Shin et al. fed a balancing
gas of water and EB in helium, over V2O5/CeO2–MgO and ach-
ieved EB conversion of 43% and ST selectivity of 91% at 600 ꢁC;
prior to the activity measurement, the catalyst was activated in
the steam ow of 50 mL minꢀ1 by heating up to 650 C at the
ꢁ
rate of 2.5 ꢁC minꢀ1
.
Sivaranjani et al. employed V-doped
12
titania with molecular oxygen at 500 ꢁC and obtained about
50% ST yield initially; however it continuously decreases at
higher time on stream.13 Various oxidants such as O2,13,14 air,15
17
N2O16 and CO2 have been used, and indeed, air as oxidant is
cost effective and preferred green way of oxidation. Above
literature reports demonstrates the wide open area of styrene
synthesis with better and sustainable yield.
Design of catalysts which are stable, gives high selectivity
(>90%) with conversion >50% for EB to ST conversion at lower
temperatures around 500 ꢁC or lower is one of the challenges in
the eld of heterogeneous catalysis.8–18 Constant efforts have
aCatalysis Division, National Chemical Laboratory, Dr Homi Bhabha Road, Pune 411
008, India. E-mail: cs.gopinath@ncl.res.in; Web: http://nclwebapps.ncl.org.in/
csgopinath; t.raja@ncl.res.in; Fax: +91-20-25902633; Tel: +91-20-25902043; +91-20-
25902006
bCentre of Excellence on Surface Science, National Chemical Laboratory, Dr Homi
Bhabha Road, Pune 411 008, India
† Electronic supplementary information (ESI) available: Experimental methods
employed and the reactor system (ESI-1). See DOI: 10.1039/c4ra11814f
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 57087–57097 | 57087

View Article Online
RSC Advances
Paper
been made to develop catalysts, operating via oxidative dehy-
drogenation (ODH) mechanism route which is also exothermic
in nature19 to lower operating temperature for EB to ST
conversion.20,21
Reaction and reactor system
The catalytic activity was evaluated using a xed bed continuous
ow reactor (FBR) having two furnace zones over a temperature
range of 430–570 ꢁC at atmospheric pressure (atm.).11,13 An
inconel reactor tube with 13 mm internal diameter and 510 mm
length was used to pack the catalyst. More detail about reactor
system is available in ref. 11 and ESI-1.†
Manganese is known for its superior catalytic properties for
various redox processes in heterogeneous catalysis. MnOx sup-
ported on silica have been also used for EB to ST conversion.22
Titania is known as excellent reducible catalytic support, but
limited with low surface area. Nonetheless, disordered meso-
porous materials23 with pseudo-three-dimensional (p3D) nature
have smaller diffusion lengths24,25 thus reactants and products
can easily diffuse to and from the active sites of disordered
mesoporous materials, which increases the selectivity and yield
of the desired product by decreasing the secondary reactions
and hence the overall rate of the reaction.
For the present manuscript, we have synthesized large
surface area Mn-doped titania (MnxTi1ꢀxO2) catalysts via solu-
tion combustion method.26–29 We used MnxTi1ꢀxO2 materials
for EB to ST conversion in a xed-bed reactor. These catalysts
were evaluated between 430 and 570 ꢁC for ST synthesis from ET
using air or oxygen as an oxidant. Present disordered meso-
porous MnxTi1ꢀxO2 materials exhibits about 57% ST yield with
high selectivity ($95%) under optimized conditions for long
durations without undergoing deactivation under reaction
condition. The present report is a part of ongoing investigations
from our group towards comprehensive understanding of metal
oxide catalysts for oxidation and ODH heterogeneous catalytic
reactions.30–38
Results and discussion
Structural and textural features
X-ray diffraction. To understand structural features of the
materials wide and low angle powder X-ray diffraction (XRD)
were recorded for MnxTi1ꢀxO2 materials. XRD results shown in
Fig. 1 exhibits diffraction features of 101, 004, 200 and 204
facets of anatase phase of titania present in MnxTi1ꢀxO2 mate-
rials. Diffraction pattern is indexed to JCPDS le no.: 21-1272,
21-1276 for anatase phase of titania with small amount of rutile
phase. About 2–9 % of rutile phase, (6.2 for TiO2, 3.8 for MT2,
8.7 for MT5, 5.8 for MT7, 7.9 for MT10, 6.8 for MT12 and 2.1 for
MT15) was also observed (# mark in Fig. 1) along with
predominant anatase phase in the fresh catalysts. No manga-
nese oxide diffraction peaks were observed up to 15% of Mn
content; indicating high dispersion of manganese ions into
TiO2 lattice. Broad peaks indicate the nanocrystalline nature of
the catalysts. Crystallite size was calculated using Debye–
Scherrer equation for all materials and shown in Table 1. Low
angle XRD recorded for all xMT materials show a single broad
Experimental section
MnxTi1ꢀxO2 (x ¼ 0.0–0.15) synthesis by combustion
Disordered mesoporous MnxTi1ꢀxO2 was prepared from cheap
chemicals and solution combustion method. All the starting
materials were obtained from spectrochem and used as such.
Manganese nitrate was used as precursor for manganese and
titanyl nitrate for titanium. Solutions were prepared with
different manganese to titanium ratio to prepare MnxTi1ꢀxO2
materials with x ¼ 0.01 to 0.15 compositions with 1 : 1 molar
ratio of urea to (Mn + Ti). Aqueous solution of precursors in 250
mL beaker was introduced into a furnace pre-heated at 400 ꢁC.
Water evaporation followed by combustion generates Mnx-
Ti1ꢀxO2 catalysts. Powder catalysts were pelletized and then
sieved for 0.8 mm size catalysts particles. Nominal Mn atom
percent is given as x in xMT for all materials.
Characterization
MnxTi1ꢀxO2 materials were characterized via various physi-
ochemical methods. Routine characterization methods are
described in our earlier manuscripts and given in ESI-1.† X-ray
photoelectron spectroscopy (XPS) measurements were per-
formed in a laboratory based custom built ambient pressure
photoelectron spectrometer (APPES from Prevac, Poland) under
UHV condition.39 XPS measurements were made with Mg Ka X-
ray source for X-ray generation and R3000HP (VG Scienta)
analyser for energy analysis.40
Fig. 1 Wide angle XRD patterns recorded from MnxTi1ꢀxO2 catalysts
with different Mn content up to 15 atom%. XRD of TiO2 prepared by
SCM is also shown for comparison. Low angle XRD pattern is given in
the inset indicates the disordered mesoporosity.
57088 | RSC Adv., 2014, 4, 57087–57097
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 1 Physicochemical characteristics of MnxTi1ꢀxO2
Catalysta
Bulk Mn content
Surface area (m2 gꢀ1
)
Pore size (nm)
Pore volume (cc gꢀ1
)
Crystallite size (nm)
TiO2
2MT
5MT
7MT
10MT
12MT
15MT
0
125
79
87
101
110
102
104
3.80
3.89
2.25
3.89
2.57
1.87
2.59
0.2281
0.1121
0.1311
0.1513
0.1811
0.0874
0.1917
8.8
7.4
8.5
8.9
7.9
6.8
7.4
1.38
4.10
6.61
7.43
11.04
14.30
a
x in xMT codes provides Mn atom%.
peak between 2q ¼ 0.8 and 1.3ꢁ (Fig. 1 – inset) indicating a elements in the materials. Representative images recorded for
disordered mesoporous nature of materials.
different elements from SEM are given in Fig. 3. Manganese and
TEM. Transmission electron microscopic images recorded titanium are present and their distribution is shown in yellow
are given for representative xMT materials in Fig. 2. TEM and blue colors in separate images, along with the particle
images exhibit particles with spherical morphology of 7–8 nm image. Homogeneous distribution of Mn in MnxTi1ꢀxO2 lattice
size, with high porosity associated with MnxTi1ꢀxO2 materials. was observed for all compositions. Mn percentage for all Mnx-
It is to be underscored that present set of disordered meso- Ti1ꢀxO2 materials were measured and shown in Table 1.
porous materials were prepared without any surfactant or
Nitrogen physisorption. Textural features of xMT materials
template molecules, compared to ordered mesoporous mate- were studied via nitrogen adsorption–desorption isotherms and
rials41–43 that employ molecules like P123 copolymer, CTAB etc. the results are shown in Fig. 4a; Barrett–Joyner–Hallenda (BJH)
as surfactant. HRTEM images exhibits d spacing for various pore size distribution plots are shown in Fig. 4b. Mesoporous
planes of TiO2. Selected area electron diffraction (SAED) nature of material can be inferred as all materials show type IV
patterns were recorded and shows predominantly 101 plane of adsorption–desorption isotherm with H2 hysteresis loop, which
anatase phases of titania. Similar porous structure was observed is typical of mesoporous materials. Surface area of material was
for TiO2, 7MT and 12MT.
calculated from Brunauer–Emmett–Teller (BET) equation, and
SEM-EDAX. Elemental mapping of MnxTi1ꢀxO2 materials shown in Table 1. BJH pore size distribution for xMT materials
were studied to understand the distribution of different exhibits bimodal (2MT and 7MT) or unimodal (all compositions
except 2MT and 7MT) pore size distribution. xMT materials
exhibits average pore diameter of around 3.5 ꢂ 0.5 nm. Calcu-
lated pore volume for MnxTi1ꢀxO2 materials are also shown in
Table 1.
Fig. 2 TEM image of (a) 2MT, (b) 5MT, (c) 10MT and (d) 15MT materials Fig. 3 (a) SEM image of a particle, and the corresponding elemental
are shown; SAED pattern is shown in inset. Spherical morphology of mapping of (b) Mn in yellow colour, and (c) Ti in blue colour through
TiO2 with (101) facets were found to be predominantly present, for all EDAX analysis on 5MT catalyst. Results indicate an uniform distribution
MnxTi1ꢀxO2 materials.
of Mn in titania.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 57087–57097 | 57089

View Article Online
RSC Advances
Paper
Fig. 4 (a) N2 adsorption–desorption isotherms, and (b) BJH pore-size
distribution of xMT catalysts.
Fig. 5 Raman spectra for xMT materials, multiplied with suitable
factors for clarity. No manganese oxide feature was observed up to
15% Mn loading, and broadening of peaks on Mn-doping indicating the
symmetry breaking of regular Raman features of anatase phase. Note
the blue shift in Eg from 145 on TiO2 (dashed line) to 159 cmꢀ1 along
with line broadening upon Mn introduction into the titania lattice.
Rutile phase features are indicated by two dotted arrows.
Spectral features
Raman spectroscopy. Raman spectral analysis of xMT
materials are shown in Fig. 5. Typical vibrational features of
anatase was observed at 145 (Eg), 198 (Eg), 398 (B1g), 516 (A1g
+
B1g) and 640 (Eg) cmꢀ1. There is shi in position of peak from
145 to 159 cmꢀ1 with an increase in Mn-content in xMT mate-
rials. Intensity of all typical features decreased drastically due to
Mn incorporation into the titania lattice. The above decrease in
the intensity of Raman features is due to the symmetry breaking
of Ti–O–Ti by Ti–O–Mn structural features due to incorporation
of Mn-ions in TiO2. XRD and TEM analysis reveals no change in
crystallinity and morphology of MnxTi1ꢀxO2 materials
compared to that of pure TiO2. Albeit its small percent of rutile
phase (Fig. 1), the same was observed with few compositions
prominently (5MT) in Fig. 5.
rather than Mn4+ oxidation state. Indeed, Mn3+ oxidation state
is likely to enhance the lattice oxygen storage/release properties
under ODH reaction conditions.
XPS. Electronic structure of xMT materials was explored via
X-ray photoelectron spectroscopy of all xMT materials. Fig. 6
shows XPS spectra of Ti2p, and Mn2p (inset) core level spectra
of xMT materials. Ti2p3/2 core level appears at a binding energy
(BE) around 459 eV for all xMT materials. This is in good
agreement with the BE reported for Ti4+ in literature reports.44
BE of Mn2p3/2 and O1s (not shown) core levels appear around
641.3 ꢂ 0.1 and 529 ꢂ 0.2 eV, respectively, for all MnxTi1ꢀxO2
materials. Observed BE of Mn2p3/2 core level around 641.3 eV
for all xMT materials indicate the oxidation state of Mn to be
3+.45,46 BE of Mn4+ state was reported to appear at 642.5 eV,
which is signicantly higher than the BE in XPS spectra, con-
rming that Mn is not present in 4 + oxidation state in xMT
3+
materials. A comparison of ionic sizes of Ti4+ (0.68 A), Mn
˚
4+
3+
˚
˚
(0.645 A) and Mn (0.53 A) also suggest the possibility of Mn
,
Fig. 6 XPS of Ti2p and Mn2p (inset) core levels of xMT materials.
57090 | RSC Adv., 2014, 4, 57087–57097
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
titania lattice. However, a comparison of Mn-content to catalytic
activity, indicates the decline in specic activity from 2 to 10%
Mn, and then it marginally increases at 12 and 15% Mn. We also
caution the readers, to take a note on changes in the catalyst
nature in the rst few hours from anatase phase xMT to Mn3O4
supported on rutile titania (vide infra). Apart from ST, benzene
is a main side product (<5%); toluene and styrene oxide are also
formed as minor products with selectivity below 2%. From the
above results, it can be inferred that Mn-doped titania centres
are the actual catalytic centres for conversion. Selectivity for ST
increases with TOS. 12MT, 10MT and 5MT showed gradual
increase, 2MT and 7MT showed steady conversion. 15MT
exhibits two fold higher yield value than TiO2. 15MT catalyst
exhibits better yield than intermediate xMT catalysts. 15MT is
the best composition among all the compositions that are
evaluated. Any further increase in doping of Mn-content resul-
ted in a mixed phase of Mn3O4 and MnxTi1ꢀxO2 indicating the
solid solubility limit of Mn in titania.
Effect of oxygen ow. O2 supply was optimized to get the
highest ST yield and selectivity. EB ow was xed at 1.8 mL hꢀ1
over xed bed of 15MT catalysts at 500 ꢁC, and O2 ow was
varied at 20, 40 and 60 mL minꢀ1 and the reaction was studied.
Measured values of conversion, yield and selectivity are plotted
for these studies, and the results are shown in Fig. 8. Conver-
sion increased with increase in oxygen ow rates. Styrene yield
increased with time due to a gradual increase in EB conversion
at 20 mL minꢀ1 O2 ow. O2 ow at 20 and 40 mL minꢀ1 aer 6 h
show constant conversion values at 35 and 50 mol%, respec-
tively, indicating the steady state attained in 5–6 h. However, 60
mL minꢀ1 O2 ow show a constant yield at 45 mol%; never-
theless, 60 mL minꢀ1 O2 ow shows a steadily declining
(increasing) selectivity (conversion). At the end of 12 h, 60 mL
minꢀ1 O2 ow exhibits 48% EB yield, and 72% styrene selec-
tivity; in contrast, at 15 h on TOS, conversion decreased to 21%
indicating the onset of deactivation (result not shown). EB
conversion increases from 25 to 35 mol% with 20 mL minꢀ1 O2
Catalytic activity
EB to styrene ODH reaction with air or molecular oxygen has
been investigated as a probe reaction to explore the catalytic
activity of xMT materials. This is primarily to explore the
inuence of Mn3+ introduced in TiO2 lattice framework.
However, caution must be exercised due to the exothermic
nature of the reaction with combustible reactant and products
in the presence of air or oxygen. Reactions were carried out at
relatively low temperature, compared to conventional endo-
thermic reaction,4,5 and at atmospheric pressure. Various
factors such as ow rates of EB, air/oxygen, catalyst composition
and temperature were varied to understand their effects on EB
conversion and selectivity to styrene.
Effect of Mn-content. For selective EB conversion to ST, and
to minimize the chances of over oxidation of easily combustible
EB and ST, active catalytic centres have to be optimized. To nd
optimum Mn-content for the highest selective oxidation, reac-
tions were carried out with different Mn-content, from TiO2 to
15 mol% Mn into TiO2 lattice. Reaction studies were carried out
with xMT materials for 12 h on time on stream (TOS) and atꢁEB
ow rate of 1.8 mL hꢀ1, and 40 mL minꢀ1 O2 ow at 500 C.
Conversion, selectivity and yield values were measured for
different xMT and are plotted in Fig. 7. Except TiO2, all xMT
catalysts show an increase in ST yield with increase in TOS. TOS
data with bare TiO2 is plotted for reference to underscore the
effect of Mn content. TiO2 exhibited about 20% ST yield, and
72–82% ST selectivity. With increase in Mn-content, EB
conversion and ST yield also increased. 15MT shows the best ST
yield (55%) and 90–95 % styrene selectivity for 12 h, and this
catalyst composition shows a sustainable activity for 45 h,
which will be discussed later. Although an increase in catalytic
activity was observed from TiO2 to 2MT, with further increasing
Mn-content the activity does not increase linearly. 2–10% Mn-
doped titania shows comparable yield. This suggests the avail-
ability of Mn on the surface is limited due to bulk doping in
Fig. 7 (a) EB conversion, (b) ST yield and (c) selectivity at 500 ꢁC are
plotted for xMT compositions. Oxygen and EB flow rate was 40 mL Fig. 8 Effect of oxygen flow on catalytic activity is shown for 20, 40
minꢀ1 and 1.8 mL hꢀ1 respectively. 15MT exhibits the highest conver- and 60 mL minꢀ1 flow. EB flow rate was maintained at 1.8 mL hꢀ1 over
sion of EB and yield of ST.
fixed bed of 15MT catalysts at 500 ꢁC.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 57087–57097 | 57091

View Article Online
RSC Advances
Paper
ow in the initial hours indicating a possibility of restructuring Measured EB conversion, styrene yield and selectivity values for
of catalyst in the transient state. 40 mL minꢀ1 O2 ow was above reactions are plotted in Fig. 10. Generally conversion
ꢁ
observed to an optimum rate for high ST yield. Benzene formed increases with increasing reaction temperature. 530 C shows
as side product along with toluene and styrene oxide with less optimum temperature for EB to ST conversion. The highest ST
than 1% selectivity. It is also to be mentioned that the gas yield was observed at 530 ꢁC with ST selectivity at 97%.
product analysis demonstrating the formation of signicant Conversion values increased marginally even up to 12 h for 500
amount of CO2 suggesting the combustion of reactant and/or and 530 ꢁC, but steady state was reached at TOS ¼ 5 h at 440 and
product, especially at high oxygen ow (Fig. 8).
470 ꢁC. 570 ꢁC demonstrates a fast increase in conversion up to
Effect of air ow. O2 was replaced by air to minimize the 8 h, followed by a decline; selectivity was also observed below
explosion hazard. Air contains diluted oxygen (21%), and hence 90% indicating the increasing contribution from combustion.
a higher ow rate is required to simulate the equivalent amount For all temperatures about 3–5% of side products, such as
of oxygen used in the results shown in Fig. 7 and 8. Due to benzene and toluene are also produced; CO2 mol% increases
diluted O2 better activity and ST selectivity was expected. from 5 to 10% conrms increasing combustion at higher
Studies at different air ow rates were carried out to understand temperatures (Fig. 10).
the effect of air ow. Air ow at 60, 120, 180 and 240 mL minꢀ1
Effect of EB ow. Rate of reactant ow over catalyst bed
were carried out at xed EB ow rate of 1.8 mL hꢀ1 over 15MTꢀa1t determines the residence time of substrate thus controls cata-
530 ꢁC, and the results are shown in Fig. 9. With 60 mL min
lysts reactivity, and hence overall product distribution. Effect of
ow there is a steady ST yield at 19%; however, the yield EB ow rates over 15MT catalysts bed were studied at air ow
increased steadily at 120 and 180 mL minꢀ1 ows. However, rate of 180 mL minꢀ1 at 530 C. To optimize EB ow, different
ꢁ
yield decreases at 240 mL minꢀ1, due to facile combustion of EB ow rates of 1.2, 1.8, 3.0 and 4.2 mL hꢀ1 were studied, and the
and ST too, towards CO2 and water. With increase in ow rates results are shown in Fig. 11. 1.2 mL hꢀ1 EB ow shows EB
the ST yield increases and the optimum yield was obtained at an conversion at 38%; however it deteriorates very fast and no
air ow rate of 180 mL minꢀ1. 180 mL minꢀ1 air ow exhibit conversion was observed at higher TOS, likely due to coke
57% ST yield $9 h with 95% selectivity. A gradual increase in deposition and hence deactivation of the catalyst. TG-DTA
conversion and yield was observed in the rst eight hours of analysis further conrms the amount of coke was about 10%
reaction at 120 and 180 mL minꢀ1 ow indicating a possible (result not shown). 1.8 mL hꢀ1 ow exhibits the highest yield of
change in the nature of catalyst. Air ow exhibits higher selec- ST; EB conversion increases gradually between 3 and 9 h, and
tivity for ST compared to O2 low, always above 95%. Mainly reaches steady state at 9 h and thereaer the reactivity was
benzene is formed (<3%) as side product.
maintained. Higher ow rates at 3 and 4.2 mL hꢀ1 shows lower,
Temperature dependence. Temperature plays an important but steady EB conversion, 97–99% ST selectivity and ST yield for
parameter for EB to ST conversion. Although EB conversion 12 h TOS. ST selectivity observed to be increasing from 90 to
increases with increase in reaction temperature, above an 97% at 1.8 mL hꢀ1 EB ow, whereas the same decreases from
optimum temperature, combustion is favoured due to favour- 90% at 1.2 mL hꢀ1 EB ow. 1.8 mL hꢀ1 ow rate provides
able oxidizing conditions; this tends to decrease the ST yield. To optimum EB ow rate for high ST yield.
optimize temperature for ST yield, studies were carried out
between 440 and 570 ꢁC at 180 mL minꢀ1 air ow rate.
Fig. 9 Air flow rate at 60, 120, 180 and 240 mL minꢀ1 were evaluated. Fig. 10 Effect of reaction temperature on activity trend, at 1.8 mL hꢀ1
EB at 1.8 mL hꢀ1 and air at 180 mL minꢀ1 flow rate on 15MT catalyst at EB flow and 180 mL minꢀ1 air flow over 15MT catalysts, between 440
530 ꢁC gives highest ST yield.
and 570 ꢁC. 530 ꢁC shows the optimum results.
57092 | RSC Adv., 2014, 4, 57087–57097
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
impregnated on TiO2 (5% Mn3O4/TiO2) catalysts powder WXRD
were also recorded and shown in Fig. 13. Narrow and high
intense peaks due to rutile phase of TiO2 at 2q ¼ 27.5 (110), 36.3
(101), 40.9 (111), 56.8 (211) are observed due to phase change of
anatase TiO2 lattice at reaction temperatures. Mn3O4 (JCPDS-
ICDD #24-0734) at 2q ¼ 37.28 and 56.90, and MnO(OH)
(JCPDS-ICDD #88-0649) (an intermediate phase of Mn3O4), at 2q
¼ 23.67, 32.21, 35.07, 37.28, 39.17, 40.125, 43.932, 60.7, 64.33,
65.45 and 69.57 are also observed due to precipitation of
manganese ions from TiO2 lattice due to reaction conditions.
Spent catalysts exhibit sharp narrow peaks, hinting a growth of
large crystallites due to aggregation. Indeed, rutile phase grow
at the cost of anatase phase, and the same is observed up to 10%
Mn; MT15 shows complete conversion of anatase to rutile
phase. Conversion of anatase to thermodynamically stable
phase,47 underscores the role of exothermic nature of the reac-
tion and its inuence in changing the nature of surface and
bulk properties of catalyst. Nonetheless, stable catalytic activity
observed for 45 h (Fig. 12) suggesting the changes cease to occur
under steady state conditions. Thus it can be inferred that
catalytic activity is due to Mn3O4 and its intermediate MnO(OH)
supported over rutile titania.
Raman spectroscopy. Raman spectroscopy analysis of spent
xMT catalysts were carried out and the results are shown in
Fig. 14. All spent catalysts exhibited a predominant rutile phase
along with minor anatase phase features of TiO2. TiO2 and 5%
Mn3O4/TiO2 catalyst's Raman spectra are given for reference.
Spent catalysts exhibits signicantly higher intensity compared
to fresh catalysts indicating the growth of crystallites to bigger
size and hence increase in crystallinity. However, no manganese
oxide features appeared in spectra of any catalysts, suggesting a
uniform distribution of them on the surface of titania.
Fig. 11 EB flow rate studied at 1.2, 1.8, 3.0 and 4.2 mL hꢀ1; with
optimized air flow at 180 mL minꢀ1 at 530 ꢁC over 15MT catalysts. 1.8
mL hꢀ1 EB flow shows the highest EB conversion and ST yield.
Catalyst stability and reaction sustainability
Optimization studies of reaction parameters for high ST yield
were demonstrated earlier and optimum values are 1.8 mL hꢀ1
EB ow, 180 mL minꢀ1 air ow at 530 C with 15MT catalyst.
ꢁ
Possible industrial application of xMT catalyst can be evaluated
by subjecting catalytic studies for longer TOS under the above
reaction conditions, to ensure robust nature of material and
obtaining constant ST yield. 15MT catalysts stability for longer
duration (45 h) was studied, and the results are shown in
Fig. 12. EB conversion increases from 45 to 58% at 15 h TOS,
and thereaer a steady 58% EB conversion was maintained. ST
yield also gradually increased from 41 to 55% and aerwards
there is no decrease in ST yield up to 45 h thus demonstrating
the sustainable nature of catalyst and reaction. Gradual
increase in catalyst activity in the rst few hours observed under
wide variety of reaction conditions demonstrates a change in
the nature of catalyst under reaction conditions towards higher
active form of the catalyst. Both conversion and selectivity
linearly increases, till the reaction reaches a steady state. ST
selectivity during whole steady state was observed to be >95%.
TG-DTA. TG-DTA of spent catalysts was measured and the
results are shown in Fig. 15. TG plots exhibits weight loss of 1–
3% between 200 and 400 ꢁC, and 1–2% weight gain between 500
Spent catalyst analysis
Aer reaction, spent catalysts, those are still active and exhibit
the high catalytic activity, were collected and analyzed with
various physiochemical techniques, such as XRD, Raman
spectroscopy, thermogravimetry and differential thermal anal-
ysis (TG-DTA), SEM, and TEM. This is mainly to explore the
nature of changes that happened to catalysts, especially under
transient state conditions which shows an induction period.
This is important, since the ODH is known to be an exothermic
reaction and local hot spots on the catalyst is likely to occur,
which could inuence the catalyst.
Fig. 12 Catalyst stability was evaluated over 45 h; EB flow at 1.8 mL hꢀ1
with optimized air flow at 180 mL minꢀ1 at 530 ꢁC over 15MT catalyst.
Catalyst showed steady ST yield for whole duration without under-
XRD. Wide angle powder XRD of spent catalysts were
analyzed to understand changes in catalysts due to catalytic
conversion of EB to ST. For comparison Mn3O4 and 5% Mn3O4 going deactivation.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 57087–57097 | 57093

View Article Online
RSC Advances
Paper
Fig. 13 Powder XRD of spent catalysts and fresh TiO2, Mn3O4 and 5%
Mn3O4/TiO2 are shown. Spent catalysts exhibits features of Mn3O4,
MnO(OH) and rutile phase of titania and this is attributed to the
exothermic nature of reaction.
Fig. 15 TG-DTA results for spent catalysts showed exothermic peaks
for coke removal up to 400 ꢁC, weight loss followed by weight gain at
700 ꢁC. Catalysts were initially subjected to the following reaction
condition at T ¼ 500 ꢁC; EB flow rate ¼ 1.8 mL hꢀ1; oxygen flow rate ¼
40 mL minꢀ1 for 12 h on TOS.
due to reaction. Other compositions tested for 12 h show
approximately the similar amount of weight loss in TGA
(Fig. 15a). In spite of some coke deposition, sustainable activity
ꢁ
observed with 15MT at 530 C demonstrates the utilisation of
air/O2 not only towards ODH reaction, but to minimize the coke
deposition. Weight gain at high temperatures is attributed to
the oxidation of manganese oxides to MnO2. Indeed, a
systematic increase in weight gain with increasing Mn-content
supports the above.
Fig. 14 Raman spectra of spent catalysts exhibits peaks due to anatase
and additional rutile phases of TiO2 supporting XRD results.
ꢁ
and 950 C; this is in addition to the initial weight loss due to
ꢁ
water removal below 200 C. DTA exhibited exothermic peaks,
which are due to carbon burning, that was deposited during
catalytic reactions over catalysts surface. MT15 catalysts tested
for 45 h shows maximum weight loss, highest peak intensity in
DTA results indicating the maximum carbon deposition of 3%
Fig. 16 (a) TEM and (b) SEM image of spent 5MT catalyst. Compared to
the TEM results shown in Fig. 2, particle agglomeration is evident. Scale
bar length is 50 nm for panel (a).
57094 | RSC Adv., 2014, 4, 57087–57097
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 18 MT15 catalysts have been tested in absence of air for catalytic
reaction at 1.8 mL EB, over 15MT catalyst at 530 ꢁC. The MT15 catalysts
regained its activity after calcination in oxygen at reaction temperature
for 20 min oxygen treatment. 20 mL minꢀ1 oxygen flow rate was
introduced at TOS ¼ 9 h.
Fig. 17 5% Mn3O4 supported on TiO2 evaluated for EB to ST under
optimized condition with air.
reaction in the absence and presence of oxygen/air and a
representative result is shown in Fig. 18. Surprisingly, 15MT
exhibited catalytic activity of about 15% EB conversion and
styrene selectivity of 95%, even in the absence of oxygen/air. A
decrease in conversion was observed from the initial 16%
conversion to 14% at TOS ¼ 8–9 h. In contrast to the general
expectation of steep decline in activity in the rst few hours,
sustenance of catalytic activity indeed indicates the supply of
lattice oxygen towards the ODH reaction. Aer demonstrating
the sustenance of reaction for nine hours, the catalyst was
SEM and TEM. To understand morphological changes with
the catalyst due to reaction, electron microscopy analysis was
conducted and the results are shown in Fig. 16. TEM results of
spent (5MT) catalyst demonstrate agglomeration to bigger
particles, compared to virgin 5MT (Fig. 2b). SEM images show
random morphology, with micron size aggregates. TEM and
SEM results are further supported by XRD analysis of spent
catalysts (Fig. 13) exhibiting sharp peaks due to bigger
agglomerated particles of spent catalyst.
calcined in oxygen (20 mL minꢀ1 ow rate) at 530 C, and EB
ꢁ
ODH with Mn3O4 supported on TiO2. In view of the above
ndings with spent catalysts, few control experiments were
carried out with Mn3O4 supported on anatase phase TiO2.
Representative result is given in Fig. 17 for 5% Mn3O4 loaded on
TiO2 by wet impregnation method and EB to ST conversion was
carried out under optimized conditions of 180 mL air per min,
EB ow of 1.8 mL hꢀ1 at 530 ꢁC. Comparable ST yield observed
in Fig. 7 for 5MT and Fig. 17 demonstrates the active nature of
catalyst is the same in both cases. In fact, the spent catalyst
analysis results are in good correlation with that of 5MT results
shown in Fig. 13 and 14. It is also to be noted that the change in
conversion and selectivity in Fig. 7 and 17 indicates the role of
gradual change in interaction between Mn3O4 and titania and
the conversion of anatase to rutile phase of titania. Although
textural properties of nanocrystalline xMT and relatively bigger
size particle of Mn3O4 on titania varies to a signicant extent,
they exhibit comparable catalytic activity indicating the minor
role of textural properties for the reaction.
conversion was continued with oxygen ow. An immediate
jump in conversion from 14 to 35% in the next three hours,
indeed, demonstrates the sustainability of the catalyst system
and ODH reaction. Above all, this observation demonstrates the
lattice oxygen role in ODH reaction. This clearly indicates xMT
catalysts do follow Mars–van Krevelen (MvK) mechanism48,49 for
EB to ST conversion.
Conclusions
Manganese incorporated in disordered mesoporous nano-
crystalline titania catalysts were prepared by simple solution
combustion method. MnxTi1ꢀxO2 materials were character-
ized via XRD, Raman spectroscopy, EDX and HRTEM tech-
niques. MnxTi1ꢀxO2 catalysts were evaluated for EB to ST
conversion by ODH route. The green chemistry approach
using air or O2 as oxidant was adopted and also it's more
economical and safer approach for the conversion. Optimum
air ow and EB ow for EB to ST conversion was found to be
180 mL minꢀ1 and 1.8 mL hꢀ1, respectively, with 15MT
Mars–van Krevelen (MvK) mechanism
Kinetics and mechanism of the reaction determines the rate composition exhibiting the highest and sustainable activity at
and distribution of products for any catalysis reaction. Different optimum reaction temperature of 530 ꢁC. Robust and
physiochemical and analytical tools helps to understand the sustainable nature of the catalyst was demonstrated by
above aspects of catalysis. In the present case, the role of lattice activity for 45 h. Conversion temperature is lowered for ST
oxygen of xMT catalyst was explored by measuring the ODH synthesis with high and sustainable yield. Catalyst is stable
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 57087–57097 | 57095

View Article Online
RSC Advances
Paper
and active for long period of time. MnxTi1ꢀxO2 materials 18 S. S. Negi, K. Sivaranjani, A. P. Singh and C. S. Gopinath,
operate via Mars–van Krevelen mechanism as conrmed by
Appl. Catal., A, 2013, 452, 132.
catalytic studies in absence/presence of oxygen.
19 Q. Wang, X. Li, W. Li and J. Feng, Catal. Commun., 2014, 50,
Spent catalyst analysis shows that active phase of catalyst is
21.
Mn3O4 supported over rutile TiO2. Indeed, there is a struc- 20 D. Sannino, V. Vaiano and P. Ciambelli, Res. Chem.
tural change occurs from anatase to rutile in the rst few Intermed., 2013, 39, 4145.
hours of reaction, due to exothermic nature of reaction. 21 N. R. Shiju, M. Anilkumar, S. P. Gokhale, B. S. Rao and
Nonetheless, the sustainable activity observed for 45 h C. V. V. Satyanarayana, Catal. Sci. Technol., 2011, 1, 1262.
demonstrates the importance of thermodynamically stable 22 R. Crǎciun and N. Dulǎmitǎ, Ind. Eng. Chem. Res., 1999, 38,
rutile phase as support with Mn3O4 as the active catalyst. 1357–1363.
Compared to the systems reported in the literature,8–21 Mn3O4 23 D. G. Kulkarni, A. V. Murugan, A. K. Viswanath and
supported over rutile TiO2 is attractive, especially in terms of
stability, and yield.
C. S. Gopinath, J. Nanosci. Nanotechnol., 2009, 9, 371.
24 T. Mathew, K. Sivaranjani, E. S. Gnanakumar, Y. Yamada,
T. Kobayashi and C. S. Gopinath, J. Mater. Chem., 2012, 22,
13484.
25 K. Sivaranjani, S. RajaAmbal, T. Das, K. Roy, S. Bhattacharyya
and C. S. Gopinath, ChemCatChem, 2014, 6, 522.
Acknowledgements
S.S.N acknowledges UGC for granting research scholarship.
Authors acknowledge the partial nancial support from CSC- 26 M. Mapa and C. S. Gopinath, Chem. Mater., 2009, 21, 351.
0404 and CSC-0125 from CSIR, New Delhi under 12FYP.
27 M. Mapa, K. Sivaranjani, D. S. Bhange, B. Saha,
P. Chakraborty, A. K. Viswanath and C. S. Gopinath, Chem.
Mater., 2010, 22, 565.
28 M. Mapa, K. S. Thushara, B. Saha, P. Chakraborty,
C. M. Janet, R. P. Viswanath, C. M. Nair, K. V. G. K. Murty
and C. S. Gopinath, Chem. Mater., 2009, 21, 2973.
Notes and references
1 M. Chanda and S. K. Roy, Industrial Polymers, Specialty
Polymers and Their Applications, CRC Press, 2008, p. 305.
2 H. A. Wittcoff, B. G. Reuben and J. S. Plotkin, Industrial 29 K. Sivaranjani and C. S. Gopinath, J. Mater. Chem., 2011, 21,
Organic Chemicals, John Wiley and sons, Hoboken, New
Jersey, 2013, p. 180.
3 H. H. Szmant, Organic Building Blocks of the Chemical
Industry, John Wiley and sons, 1989, p. 431.
2639.
30 S. Velu, N. Satoh, C. S. Gopinath and K. Suzuki, Catal. Lett.,
2002, 82, 145.
31 S. Velu, K. Suzuki and C. S. Gopinath, J. Phys. Chem. B, 2002,
106, 12737.
4 M. Muhler, J. Schutze, M. Wesemann, T. Raymant, A. Dent,
¨
R. Schlogl and G. Ertl, J. Catal., 1990, 126, 339.
32 M. K. Dongare, V. Ramaswamy, C. S. Gopinath,
A. V. Ramaswamy, S. Scheurell, M. Brueckner and
E. Kemnitz, J. Catal., 2001, 199, 209.
5 G. P. Chiusoli and P. M. Maitlis, Metal-catalysis in industrial
organic processes, 2008, p. 109.
6 Y. She, J. Han and Y. H. Ma, Catal. Today, 2001, 67, 43.
7 J. C. S. de Araujo, C. B. A. Sousa, A. C. Oliveira, F. N. A. Freire,
A. P. Ayala and A. C. Oliveira, Appl. Catal., A, 2010, 377, 55.
8 C. Nederlof, V. Zarubina, I. Melian-Cabrera, E. H. J. Heeres,
33 S. Velu, K. Suzuki, M. Vijayaraj, S. Barman and
C. S. Gopinath, Appl. Catal., B, 2005, 55, 287.
34 N. R. Shiju, M. Anilkumar, S. P. Mirajkar, C. S. Gopinath,
C. V. Satyanarayana and B. S. Rao, J. Catal., 2005, 230, 484.
F. Kapteijn and M. Makkee, Catal. Sci. Technol., 2013, 3, 519. 35 T. Mathew, Y. Yamada, A. Ueda, H. Shioyama, T. Kobayashi
9 N. V. Qui, P. Scholz, T. F. Keller, K. Pollok and
B. Ondruschka, Chem. Eng. Technol., 2013, 36, 300.
10 R. Watanabe, M. Ikushima, K. Mukawa, F. Sumomozawa,
S. Ogo and Y. Sekine, Front. Chem., 2013, 21, 1.
11 A. K. Venugopal, A. T. Venugopalan, P. Kaliyappan and
T. Raja, Green Chem., 2013, 15, 3259.
12 V. K. Nguyen, J. H. Park and C. H. Shin, Korean J. Chem. Eng.,
2014, 31, 582.
13 K. Sivaranjani, A. Verma and C. S. Gopinath, Green Chem.,
2012, 14, 461.
and C. S. Gopinath, Appl. Catal., A, 2006, 300, 58.
36 P. Maity, C. S. Gopinath, S. Bhaduri and G. K. Lahiri, Green
Chem., 2009, 11, 554.
37 T. Rajesh, A. K. Rajarajan, C. S. Gopinath and R. N. Devi, J.
Phys. Chem. C, 2012, 116, 9526.
38 A. C. S. Sekhar, K. Sivaranjani, C. S. Gopinath and
C. P. Vinod, Catal. Today, 2012, 198, 92.
39 (a) K. Roy, C. P. Vinod and C. S. Gopinath, J. Phys. Chem. C,
2013, 117, 4717; (b) K. Roy and C. S. Gopinath, Anal. Chem.,
2014, 86, 3683.
14 C. Nederlof, V. Zarubina, I. V. M. Cabrera, E. H. J. Heeres, 40 P. Devaraji, N. K. Sathu and C. S. Gopinath, ACS Catal., 2014,
F. Kapteijn and M. Makkee, Appl. Catal., A, 2014, 476, 204.
4, 2844.
15 N. T. Thao and H. H. Trung, Catal. Commun., 2014, 45, 153. 41 N. Maity, P. R. Rajamohanan, S. Ganapathy, C. S. Gopinath,
16 (a) P. Kustrowski, L. Chmielarz, R. Dziembaj, P. Cool and
S. Bhaduri and G. K. Lahiri, J. Phys. Chem. C, 2008, 112, 9428.
E. F. Vansant, J. Phys. Chem. A, 2005, 109, 330; (b) 42 A. Lazar, W. R. Thiel and A. P. Singh, RSC Adv., 2014, 4,
K. Thirunavukkarasu, K. Thirumoorthy, J. Libuda and
C. S. Gopinath, J. Phys. Chem. B, 2005, 109, 13272.
17 M. C. Rangel, A. P. M. Monteiro, S. G. Marchetti, S. B. Lima
and M. S. Ramos, J. Mol. Catal. A: Chem., 2014, 387, 147.
14063.
43 E. S. Gnanakumar, J. John, T. Raja and C. S. Gopinath, J.
Nanosci. Nanotechnol., 2013, 13, 2682.
57096 | RSC Adv., 2014, 4, 57087–57097
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
44 M. Satish, R. P. Viswanath and C. S. Gopinath, J. Nanosci. 47 (a) C. S. Gopinath and T. Raja, J. Phys. Chem. B, 2001, 105,
Nanotechnol., 2009, 9, 423.
45 S. Bag, K. Roy, C. S. Gopinath and C. Retna Raj, ACS Appl.
Mater. Interfaces, 2014, 6, 2692.
12427; (b) B. Naik, K. M. Parida and C. S. Gopinath, J. Phys.
Chem. C, 2010, 114, 19473.
48 P. Mars and D. W. van Krevelen, Chem. Eng. Sci., 1954, 3, 41.
46 B. Murugan, D. Srinivas, C. S. Gopinath, V. Ramaswamy and 49 C. Doornkamp and V. Ponec, J. Mol. Catal. A: Chem., 2000,
A. V. Ramaswamy, Chem. Mater., 2005, 17, 3983.
162, 19.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 57087–57097 | 57097