Applied Catalysis A, General 559 (2018) 175–186

Contents lists available at ScienceDirect

Applied Catalysis A, General

journal homepage: www.elsevier.com/locate/apcata

Excellent product selectivity towards 2-phenyl-acetaldehyde and styrene T

oxide using manganese oxide and cobalt oxide NPs for the selective
oxidation of styrene
⁎
Ngonidzashe Masungaa, Bryan P. Doyleb, Emanuela Carleschib, Reinout Meijbooma,
a
Department of Chemistry, University of Johannesburg, P.O. Box 524, Johannesburg, South Africa
b
Department of Physics, University of Johannesburg, P.O. Box 524, Auckland Park 2006, Johannesburg, South Africa

A R T I C LE I N FO A B S T R A C T

Keywords: MnOx and Co3O4 nanoparticles (NPs) were synthesized using the temperature controlled co-precipitation
Manganese oxide nanoparticles method and were immobilized on the mesoporous metal oxide supports that were synthesized using the inverse
Cobalt oxide nanoparticles surfactant micelle method. The synthesized catalysts were characterized using TEM, SEM, pXRD, N2-sorption,
Mesoporous metal oxides XPS, and H2-TPR. From TEM the well-dispersed metal oxide NPs were observed and the interaction of the NPs
Styrene oxide
with the support was confirmed with H2-TPR. These metal oxide NPs and immobilized metal oxide NPs were
2-Phenyl-acetaldehyde
employed in the selective oxidation of styrene using tert-butyl hydrogen peroxide (TBHP) as an oxidant. Higher
Styrene oxidation
selectivity towards styrene oxide was obtained using Co3O4 NPs. MnOx NPs gave higher selectivity towards 2-
phenyl-acetaldehyde while the immobilized metal oxide NPs gave a mixture of products. The catalytic activities
of the compared catalysts followed the following decreasing order of MnOx NPs (646.9 h−1) > Co3O4 NPs
(338.1 h−1) > Co-MnO2_350 (54.1 h−1) > Mn-Co3O4_350 (44.1 h−1) ˃ MnO2_350 (35.8 h−1) > Co3O4_350
(21.3 h−1) after 8 h of reaction. The catalytic activities and selectivity profiles under various reaction conditions
such as solvent type, catalyst amount, styrene to TBHP ratio and temperature are discussed. The catalytic re-
cyclability of immobilized metal oxide NPs on mesoporous metal oxides showed that they are stable, and the
structure of the catalyst was retained.

1. Introduction and benzoic acid are a result of further oxidation of styrene oxide while
2-phenyl-acetaldehyde and acetophenone are a result of the iso-
The development of heterogeneous catalysts has been an area of merization of styrene oxide. The schematic illustration of styrene oxi-
great interest over the past decades, due to the ease of recycling of these dation is given in Scheme 1.
heterogeneous catalysts. However, there have been great strides in the Styrene oxide is mainly used in the epoxy resins industry as a
field of catalysis such as the successful drive to move from the tradi- plasticizer and also, as a raw material for phenethyl alcohol, styrene
tional complex compounds to nanoscale materials [1]. The ability to glycol, and its derivatives. It is also used as a chemical intermediate for
prepare NPs with well-defined shapes, with sizes that range between cosmetics, surface coatings, treatment of fibers and textiles, and agri-
1–100 nm in diameters, have attracted great attention from researchers cultural and biological chemicals and is, therefore, an important pro-
[2]. These NPs have unique properties such as electronic, magnetic and duct for the pharmaceutical and fine chemical industry [9–11]. While,
catalytic properties that distinguish them from their corresponding bulk 2-phenyl-acetaldehyde, the isomer of styrene oxide, is mainly used in
materials [2–4]. the agricultural industry for monitoring and controlling insects because
The selective oxidation of styrene to its products has been much of its aroma that attracts them. It is also used as an intermediate in the
studied in the past decades as its products are of industrial importance biochemical pathways, including those involved in the metabolism of
[5]. It has been also used as a model reaction for ethylene epoxidation carbohydrates, vitamins, steroids, amino acids, benzylisoquinoline al-
due to it being a liquid at room temperature [3]. Styrene oxidation kaloids, hormones, and lipids [12]. Therefore, it is important to selec-
produces multiple products such as styrene oxide, benzaldehyde, 2- tively oxidize styrene to styrene oxide and 2-phenylacetaldehyde while
phenyl-acetaldehyde, acetophenone, benzoic acid [3,6–8], depending achieving high styrene conversion.
on the reaction conditions and the type of catalyst used. Benzaldehyde We intend to exploit the properties of manganese oxide (MnOx) NPs

⁎
Corresponding author.
E-mail addresses: [email protected], [email protected] (R. Meijboom).

https://doi.org/10.1016/j.apcata.2018.04.027
Received 18 January 2018; Received in revised form 20 April 2018; Accepted 21 April 2018
Available online 23 April 2018
0926-860X/ © 2018 Elsevier B.V. All rights reserved.
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Scheme 1. Schematic illustration of the catalytic oxidation of styrene.

Fig. 1. HR-TEM images of (A): MnOx NPs, (C): Co3O4 NPs, and their particle size distribution histograms in (B) and (D) respectively.

and cobalt oxide (Co3O4) NPs which include high surface to volume [21]. The MnOx NPs have also been used in hydrogenation reactions
ratio and excellent processability to selectively produce styrene oxide such as reduction of carbon dioxide to methanol [22] and also as
and 2-phenylacetaldehyde. From literature, it can be noted that dif- electrocatalysts for the oxygen reduction reaction in alkaline medium
ferent forms of manganese oxide catalysts have been employed in [23]. However, in this work, the manganese oxide nanoparticles which
styrene oxidation and different products were observed. Benzaldehyde were synthesized using the heat mediated co-precipitation method was
and styrene oxide were the major products observed using manganese investigated for the oxidation of styrene.
oxide catalytic systems [13–18]. For example, the role of manganese Different forms of cobalt catalysts which include cobalt complex,
oxide octahedral molecular sieves in styrene epoxidation [19], man- cobalt nanocrystals, cobalt oxides, etc, have also been employed in the
ganese catalyzed epoxidation of alkenes in bicarbonate solutions [20], oxidation of styrene to styrene oxide and benzaldehyde as major pro-
etc. The Mn3O4 and Co3O4 nanocrystals which were synthesized using ducts [7,16,17,24,25]. However, cobalt oxide NPs synthesized using the
the sonochemical method were used for the epoxidation of styrene heat mediated co-precipitation method has not been used for studying

176
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 2. TEM images A: Mn-Co3O4_350 and B: Co-MnO2_350.

Table 1 2.2. Synthesis of manganese oxide NPs (MnOx)

The table of BET surface areas, BJH pore sizes, pore volumes and crystallite
sizes of the synthesized mesoporous metal oxides catalysts. Manganese oxide NPs were synthesized using Mn-oleate as a source
Catalyst BET Surface BJH pore Pore Volume Crystallite sizea of manganese. Mn-oleate was synthesized by dissolving 7.92 g
areas (m2/g) size (nm) (cm3) (nm) (40 mmol) MnCl2.4H2O and 24.36 g (80 mmol) sodium oleate in a
mixture of 30 mL of ethanol, 40 mL of water, and 70 mL n-hexane that
MnO2 _ 350 37 22.4 0.21 11.5
was preheated to 70 °C. The mixture was left to react for 4 h at 70 °C
Co-MnO2_350 36 22.1 0.20 12.0
Co3O4 _ 350 42 26.4 0.27 20.8 under magnetic stirring. The solution of Mn-oleate was collected using
Mn-Co3O4 _ 350 38 25.9 0.32 19.5 a separation funnel. Then, the Mn-oleate salt was collected using a
rotary evaporator.
a
Crystallite sizes were calculated from p-XRD using the Scherer equation. The MnOx NPs were synthesized by dissolving 1.24 g of Mn-oleate in
12.67 mL of 1-octadecene in a 50 mL round bottom flask. The solution
the oxidation of styrene. Cobalt NPs have been used as an efficient was stirred at 70 °C and degassed under nitrogen for 1 h. The tem-
electrocatalyst for oxygen evolution reactions [26,27]. Other transition perature of the reaction mixture was ramped to 260 °C (1.9 °C/min).
metal NPs that have been investigated in styrene oxidation include Pt The color of the solution changed from pink to dark brownish. The
and Pd by Noh [3] and supported gold NPs [2,28,29]. Therefore, it is solution was left to cool for 10 min. The MnOx NPs were collected by
important to investigate the behavior of our unsupported MnOx and centrifugation and they were dispersed in 10 mL of chloroform [22,30].
Co3O4 NPs in styrene oxidation. Also, supporting these NPs on an active
metal oxide support will enhance our knowledge on the influence of the
created interface between the support and the metal NPs. The MnOx 2.3. Synthesis of cobalt oxide NPs
NPs were supported on mesoporous cobalt oxide, while Co3O4 NPs was
supported on mesoporous manganese oxide to investigate the influence Cobalt oxide NPs were synthesized by degassing 150 μL of oleic acid
of the interfacial domain that is created by the change in electronic under nitrogen for 1 h that was placed in a Schlenk tube connected to a
structure through the exchange of electrons between the two metals condenser. Then, 12 ml of dichlorobenzene (DCB) was transferred into
[22]. Hence, exploration of the catalytic behavior of MnOx and Co3O4 the flask and the temperature was raised to 169 °C (5 °C/min). Finally,
NPs in catalytic oxidation of styrene in this study will enrich the sci- 0.444 g of Co2(CO)8 dissolved in 3 mL DCB was injected into a reaction
entific knowledge in the field of oxidation by colloidal metal oxide NPs. vessel using an airtight syringe and the reaction was left to react for
20 min before the heat was turned off. DCB (13 mL) was poured into the
2. Experimental details flask, and the colloidal suspension was left to cool. The Co3O4 NPs were
collected by centrifugation and redispersed in 10 mL chloroform for
2.1. Materials and instrumentation further use [22].

All chemicals were used as received without further purification and

were of the analytical grade. Styrene (≥99%), TBHP (70% in water), 2.4. Immobilization of NPs on mesoporous metal oxides
styrene oxide (≥97%), benzaldehyde (≥99%), acetophenone (≥99%),
2-phenyl-acetaldehyde (≥98%), dichloromethane (DCM) (≥99.8%), Immobilization of the NPs was achieved using the capillary inclu-
dimethyl carbonate (DMC) (≥99%), methanol (≥99.5%), cyclohexane sion method. Briefly, 5 mL (0.570 mmol) of colloidal MnOx NPs dis-
(≥99.8%), 1-butanol (≥99.5%), dichlorobenzene (≥99%), ethanol persed in chloroform was diluted with 40 mL n-hexane and mixed with
(≥99.8%), n-hexane (≥97%), 1-octadecene (≥98%), toluene 1 g of mesoporous cobalt oxide. The mixture was sonicated for 3 h at
(≥99.8%), chloroform (≥99%), and acetonitrile (≥99.9%), manga- room temperature. The precipitate was collected by centrifugation and
nese(II) nitrate tetrahydrate (≥97%), cobalt(II) nitrate hexahydrate washed with a 50/50 v/v mixture of ethanol and acetone five times.
(≥98%), MnCl2·4H2O (≥98%), dichlorobenzene (DCB) (≥99%) so- The resultant pellet was dried overnight at 80 °C. The same procedure
dium oleate (≥82%), Co2(CO)8 (stabilized with 1–5% hexanes), oleic was followed for immobilizing Co3O4 NPs on mesoporous manganese
acid (≥ 99%), and poly (ethylene glycol)-block-poly (propylene oxide [22]. The synthesis of mesoporous metal oxides is detailed in the
glycol)-block-poly (ethylene glycol) (P123) were all purchased from supplementary information. The description of the instruments used for
Sigma-Aldrich. characterization is found in the supplementary information.

177
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 3. Characterization of immobilized metal oxide NPs on MMOs. A: N2 sorption isotherms; B: BJH pore volume distributions, and C: BJH pore sizes distributions.

Fig. 4. Characterization of immobilized metal oxide NPs mesoporous metal oxides. A: graphs of low angle p-XRD and B: graphs of wide angle p-XRD patterns.

2.5. Experimental procedure for styrene oxidation silicone oil bath was heated to 90 °C prior to the reaction and the re-
actions were heated under reflux with stirring. The reaction was sam-
The oxidation of styrene was conducted under aerobic conditions pled after every 2 h for 8 h. The catalytic conversion of styrene and
using a 25 mL 2-neck round bottom flask equipped with a magnetic selectivity was determined using a Shimadzu GC-2010 Plus equipped
stirrer, thermometer, and a heated silicon oil bath. For colloidal MnOx with a flame ionization detector (FID) using a Restek - 800-356-1688
and Co3O4 NPs, 0.114 mmol of each was used. For immobilized MnOx capillary column (30 m * 0.25 mm, the film thickness of 0.25 μm).
and Co3O4 NPs on mesoporous metal oxides, typically 0.1 g of catalyst, Temperatures of 250 °C and 300 °C were used for injection port and FID
1 mmol styrene, and 2 mmol TBHP as an oxygen source was added to a respectively. The column oven temperature was set at 300 °C and ni-
25 mL 2-neck round bottom flask filled with 5 mL DMC as a solvent. The trogen was used as a carrier gas. Reaction products were confirmed

178
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 5. H2-TPR profiles of mesoporous metal oxides and when immobilized with metal oxide NPs. A: cobalt oxide and B: manganese oxide.

using a Shimadzu gas chromatography (GC–MS) using a Rxi-5Sil MS NPs. Fig. 3A shows that both immobilized mesoporous metal oxides
capillary column (30 m * 0.25 mm ID * 0.25 μm film thickness) and exhibited the characteristic type IV adsorption isotherms, with a hys-
injection port temperature of 200 °C was used. The ion source and in- teresis loop of the ordered mesoporous structure.
terface temperature were 200 °C and 250 °C, respectively. From Fig. 4A, it was observed that the synthesized materials diffract
in a low angle region. The peak observed at around 2θ showed that both
immobilized metal oxide NPs on mesoporous metal oxides are crystal-
3. Results line in nature. The crystallite sizes increase did not change significantly
after immobilization of the metal oxide NPs as shown in Table 1. These
3.1. Synthesis and characterization crystallite sizes were calculated from wide angle p-XRD shown in
Fig. 4B, using the Scherrer equation. Fig. 5 shows the H2-TPR profiles of
The metal oxide NPs were successfully synthesized using the tem- mesoporous metal oxides and when immobilized with metal oxide NPs
perature-controlled co-precipitation method. The synthesized MnOx and it can be noted that two bands are observed. It is observed that
NPs were well dispersed as noted from the TEM image in Fig. 1A and immobilization of metal oxide NPs on the mesoporous metal oxide
were found to have an average size of 6.2 nm as shown in Fig. 1B. The support increases the reduction temperature of the mesoporous metal
synthesized Co3O4 NPs were also well dispersed, and they were found oxides.
to have an average size of 8.4 nm. The purity of the metal oxide NPs X-ray photoelectron spectroscopy (XPS) was carried to provide in-
was confirmed using EDX as shown in Fig. S1. From Fig. S1A, the formation on the electronic structure of the catalysts. The O 1s spectra
manganese and oxygen peaks were observed in the EDX spectrum for all four catalysts are as expected [32]. The largest peak in Fig. 6, at
which confirms the presence of pure MnOx NPs. The copper peaks the lowest binding energy, is due to stoichiometric oxygen. The peak at
which were observed came from the Cu grid. In Fig. S1B, the cobalt and around 531 eV is due to oxygen vacancies and/or other defects [33,34].
oxygen peaks were observed which confirms the presence of Co3O4 NPs Finally, the peak at highest binding energy is known to be from che-
The elemental analysis of immobilized metal oxide NPs on meso- misorbed oxygen [35]. In a similar way the Mn 2p spectra, shown in
porous metal oxide supports was confirmed using EDX as shown in Fig. Fig. 6, show spin-orbit split peaks. These are due to Mn in a 2 + oxi-
S1. From the ICP-OES analysis, 1.2 wt% of Mn was found to be sup- dation state, a 3 + oxidation state and a satellite at high binding en-
ported on mesoporous cobalt oxide and 1.2 wt% of Co was found to be ergy. Note that for Mn the 2 + oxidation state is at lower binding en-
supported on mesoporous manganese oxide support. Fig. 2 shows the ergy, which is the opposite to what occurs with Co [36]. The oxidation
TEM images of the immobilized NPs on the metal oxides support. It is state % ratio of Mn 2p are presented in Table 2. It can be noted that the
important to note the difficulty in observing the supported NPs on the two samples with MnO2 have similar oxidation states, whereas the
support and measuring them since the NPs suffer from low contrast due MnOx NPs samples have a slightly different oxidation state. Fig. 6 shows
to the high electron density from the support. The difference in focal the Co 2p spectra of those catalysts with Co present. The spectra were
planes of the NPs and the support when acquiring images adds to the fitted with three components. Each component is split into the Co 2p3/2
difficulty of measuring the NPs [31]. In Fig. 2A and B) the small black and 2p1/2 peaks. These components are due to Co in the 3 + oxidation
dots are observed on the support which signifies the presence of the state (at lower binding energy), Co in the 2 + oxidation state (at higher
MnOx NPs and Co3O4 NPs on the support respectively. The SEM images binding energy) and a satellite at high binding energy. The spectra are
in Fig. S2A and B show the morphology of the immobilized MnOx NPs consistent with what is expected from literature [32]. The oxidation
and Co3O4 NPs on the support respectively. state % ratio of Co 2p are presented in Table 2, and they were found to
The nitrogen sorption experiments showed that the BJH pore sizes be similar across all 3 samples.
decreased from 22.4 nm for mesoporous MnO2_350 to 22.1 nm after the
Co3O4 NPs were immobilized on it. For mesoporous Co3O4_350 it de-
creased from 26.4 to 25.9 nm after MnOx NPs were immobilized, as 3.2. Optimization of the reaction conditions
shown in Table 1. The surface area for mesoporous MnO2_350 de-
creased from 37 to 36 m2/g after immobilization of Co3O4 NPs. While The oxidation of styrene in the presence of air using metal oxide NPs
the surface area of Co3O4_350 decreased from 42 to 38 m2/g after im- and immobilized metal oxide NPs on mesoporous metal oxides support
mobilization of MnOx NPs. These changes in the physical properties of as catalysts were conducted with various experimental parameters such
the mesoporous metal signify successful immobilization of metal oxide as solvent type, catalyst amount, etc, to obtain the optimum conditions.

179
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 6. The O 1 s, Mn 2p, and Co 2p core level electrons XPS spectra for pure mesoporous metal oxides and for immobilized metal oxide nanoparticles on a
mesoporous metal oxide support.

Table 2 phenyl-acetaldehyde was lower (81.6%) than that of DMC (90.6%) as

The percentage composition of the oxidation states of the synthesized catalysts. shown in Fig. 7A. From Fig. 7A, it is noted that DMC as a solvent gave
Orbital type Catalyst Oxidation state (2+) Oxidation state (3+)
high selectivity (90.6%) towards 2-phenyl-acetaldehyde while its
% % styrene conversion is moderate as compared to others. Solvents with a
high polarity index such as acetonitrile (polarity index 5.8) and me-
Co 2p Co3O4 _350 28 72 thanol (polarity index 5.1) showed low styrene conversion of 50.9%
Mn-Co3O4 _350 21 79
Co-MnO2_350 24 76
and 38.7% respectively while their product selectivity was moderate
Mn 2p MnO2_350 66 34 compared to others. Toluene and cyclohexane which are nonpolar
Co-MnO2_350 62 38 solvents with a polarity index of 2.4 and 0.2 respectively gave a rela-
Mn-Co3O4 _350 53 47 tively high styrene conversion and a moderate selectivity towards 2-
phenyl-acetaldehyde as shown in Fig. 7A. After 4 h of reaction, small
brown particles were observed in methanol while in acetonitrile, large
3.2.1. Solvent type
black particles were observed. The overall results showed that DMC as a
The influence of solvent on the oxidation of styrene was investigated
solvent gave the highest product selectivity and a moderate styrene
using various types of solvents. Chloroform was found to give the
conversion, hence it was selected as an optimum solvent for further
highest styrene conversion of 67.8%, however, its selectivity towards 2-

180
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 7. Optimization of the reaction conditions. A: solvent variation; B: catalytic amount variation; C: mole ratio of styrene to TBHP variation; D: temperature
variation. Reaction conditions: catalyst (0.114 mmol of MnOx NPs), solvent (5 mL DMC), 1 mmol styrene, 2 mmol TBHP, T = 90 °C, time = 4 h, except when the
condition is investigated.

investigations. 3.2.4. The optimum temperature

The influence of temperature on the styrene conversion and product
selectivity was investigated. Styrene conversion was found to increase
3.2.2. Optimum amounts of reagents
with increase in temperature. The highest selectivity towards 2-phenyl-
The influence of the catalyst amount on the styrene conversion and
acetaldehyde (91.7%) was achieved at 80 °C. From the observed results,
product selectivity was investigated and it was observed that the
no correlation can be deduced between temperature and product se-
styrene conversion and product selectivity towards 2-phenyl-acet-
lectivity as the product selectivity remains relatively constant as shown
aldehyde increase as the catalyst amount is increased as shown in
in Fig. 7D. Therefore, 90 °C was used as the reaction temperature for
Fig. 7B. The reaction was performed without the catalyst under the
further investigations on styrene oxidation considering the boiling point
same reaction conditions and a conversion of 8.5% was observed. When
of the solvent and styrene conversion.
the catalyst amount was increased beyond 0.114 mmol, a large decrease
in selectivity towards 2-phenyl-acetaldehyde was observed. Hence
0.114 mmol was chosen as the amount of catalyst for further research
3.3. Styrene oxidation catalyzed by various catalysts
on styrene oxidation.
From Fig. 8 it can be noted that styrene conversion increases with an
3.2.3. The optimum ratio of styrene to TBHP increase in reaction time for all catalysts. The highest styrene conver-
The influence of styrene to TBHP ratio in the styrene conversion and sion (59.3%) was obtained using MnOx NPs as a catalyst after 8 h. The
product selectivity was examined by keeping the concentration of catalytic activities of the compared catalysts were decreasing in the
styrene constant while varying the amount of TBHP. From Fig. 7C, it is order of MnOx NPs (646.9 h−1) > Co3O4 NPs (338.1 h−1) > Co-
noted that the styrene conversion increases with an increase in TBHP MnO2_350 (54.1 h−1) > Mn-Co3O4 _ 350 (44.1 h−1) ˃ MnO2_350
amount. Also, the selectivity towards 2-phenyl-acetaldehyde was af- (35.8 h−1) ˃ Co3O4 _ 350 (21.3 h−1) after 8 h of reaction. The turnover
fected to a lower extent. However, a large decrease in selectivity to- frequency (TOF), which is defined as the number of moles of styrene
wards 2-phenyl-acetaldehyde was observed when the styrene to TBHP converted to all products per mole of the catalyst per hour were cal-
ratio is at 1:4 as shown in Fig. 7C. Hence, the ratio of 1:1 was found to culated after 8 h of reaction. When the mesoporous metal oxide sup-
be the optimum condition for further research on the styrene oxidation. ports were used in styrene oxidation as catalysts, poor styrene

181
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 8. The comparison of styrene conversion and product selectivity for different metal NPs and after being immobilized on mesoporous metal oxide supports. A:
MnOx NPs, B: Co3O4 NPs, C: Mn-Co3O4_350, D: Co-MnO2_350, and E: turnover frequency. Reaction conditions: catalyst (1 mL for metal oxide NPs and 0.1 g for
immobilized metal oxide NPS), solvent (5 mL DMC), 1 mmol styrene, 1 mmol TBHP, T = 90 °C.

conversion and product selectivity were observed. From Fig. 8E it can observed as a major product while when colloidal Co3O4 NPs were
also be noted that immobilizing metal oxide NPs on the mesoporous used, styrene oxide was obtained. When these metal oxide NPs were
metal oxides supports reduces the activity of the metal oxide NPs in immobilized on the mesoporous metal oxides support they lost their
styrene oxidation. From the obtained results, it can be noted that the selectivity and a mixture of products were obtained as shown in Fig. 8
selectivity of the products formed during oxidation of styrene depends (C and D). For metal oxide NPs, the product selectivity remains constant
on the catalyst used. towards one product over the course of the reaction from 2 to 8 h.
When colloidal MnOx NPs were used, 2-phenyl-acetaldehyde was However, with immobilized metal oxide NPs, the product selectivity

182
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 9. The influence of catalyst recycles test on styrene conversion and product selectivity. A: Mn-Co3O4_350 and B: Co-MnO2_350. Reaction conditions: 100 mg
catalyst, Solvent (5 mL DMC), 1 mmol styrene, 1 mmol TBHP, T = 90 °C, Time = 4 h.

changed with the change in time. When Mn-Co3O4_350 was used as a techniques. From the nitrogen, sorption experiments the observed Type
catalyst, the formation of 2-phenyl-acetaldehyde increased while the IV adsorption isotherms signify the regular ordered mesoporous nature.
formation of styrene oxide decreased with an increase in reaction time. The presence of the regular ordered mesoporous structure was sup-
For the Co-MnO2_350 catalyst, the formation of other products in- ported by the presence of a low angle peak observed in the p-XRD [38].
creases with increase in reaction time while product selectivity towards After immobilization of metal oxide, NPs on the mesoporous metal
styrene oxide and 2-phenyl-acetaldehyde decreased from 4 to 8 h. oxides, the surface area, and pore volumes decreased. This is because
Hence, immobilizing metal oxide NPs on mesoporous metal oxide the metal oxide NPs occupy space inside pores. [40]. The purity of the
supports decreased both styrene conversion and product selectivity. metal oxide NPs and immobilized metal oxide NPs on the support was
Representative GC–MS spectra of 2-phenyl-acetaldehyde and styrene confirmed by EDX analysis which showed only the peaks of oxygen and
oxide are given in Figs. S3 and S4, respectively of the supplementary the corresponding metals under investigation as shown in Fig. S1.
materials. The H2-TPR was performed to investigate the differences in redox
properties of the immobilized metal oxide NPs on mesoporous metal
oxide supports because the catalytic activity of the catalyst towards
3.3.1. Catalyst recycle test
redox reactions depends on the reducibility. For spinel metal oxides,
The influence of catalyst recycling on styrene conversion and pro-
usually, the first reduction signal is assigned to the most electropositive
duct selectivity was investigated by recycling the catalysts four times
cation. The first band appearing at 268 °C for mesoporous cobalt oxides
for immobilized metal oxide NPs. From Fig. 9, it can be noted that
can be ascribed to the reduction of Co3+ species to Co2+ while the
styrene conversion diminished, and product selectivity showed insig-
second band at 348 °C can be ascribed to the reduction of Co2+ to
nificant changes even after the 4th cycle for Co-MnO2_350. Mn-
metallic cobalt [37,39]. For mesoporous manganese metal oxide, the
Co3O4_350 showed higher selectivity towards 2-phenylacetaldehyde in
first band appearing at 272 °C can be ascribed to the reduction of Mn3+
the first two cycles and then, the selectivity changed towards styrene
to Mn2+ while the second band 405 °C can be ascribed to the reduction
oxide for the third and fourth cycles. The immobilized metal oxide NPs
of Mn2+ to metallic manganese [38,41]. The obtained oxidation states
on mesoporous metal oxide catalysts remained unchanged as shown in
are supported by the data obtained from the XPS measurements shown
the TEM images (Fig. 10E and F). The TEM images taken after 8 h of the
in Table 2. From Fig. 5, the increase in reduction of the mesoporous
reaction showed slight agglomeration for Co3O4 NPs while most of the
metal oxide supports after immobilization of the metal oxide NPs can be
NPs remain well dispersed in solution as seen in Fig. 10A. The Co3O4
attributed to the change in the electronic structure of the material due
NPs increased in size from 8.4 to 9.7 nm during the reaction. The shape
to the interaction of the NPs with the support [22].
of MnOx NPs remained unchanged after the reaction as shown in
Fig. 10C and their particle size also increases from 6.2 to 7.8 nm after
the reaction. This showed the stability of the metal oxide NPs in the
4.2. Selective oxidation of styrene
reaction.
4.2.1. Optimization of the reaction conditions: solvent type, catalyst
4. Discussion amount, the ratio of styrene to TBHP and temperature variation
In literature, the influence of solvent type used for the oxidation of
4.1. Synthesis and characterization of colloidal NPs and MMOs styrene was found to influence the styrene conversion and the product
immobilized with metal oxide NPs selectivity of the reaction. Higher styrene conversion was reported to be
obtained for solvents with a high polarity such as acetonitrile, DMF etc
The mesoporous metal oxide supports used to immobilize metal [17,18,24,42,43]. Therefore, considering the results of these papers,
oxide NPs in this study were synthesized using the inverse micelle sol- polar solvents and non-polar solvents were investigated. The lower
gel method. The principle behind this method is that it uses HNO3 to styrene conversion obtained using acetonitrile and methanol as solvents
prevent condensation while the metal oxo-clusters are stabilized and can be attributed to the agglomeration of NPs in solution which results
the clusters are formed inside hydrated inverse micelles and they in- in them being inactive catalysts. DMC which is also a polar solvent with
teract with the surfactant via hydrogen bonding [37–39]. a polar index of 3.9 gave higher 2-phenylacetaldehyde selectivity and
The synthesized materials were characterized using a range of moderate styrene conversion. The good dispersion of NPs in DMC and

183
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Fig. 10. TEM images were taken after the recycle test of styrene oxidation: A: Co3O4 NPs, C: MnOx NPs E: Mn-Co3O4_350, F: Co-MnO2_350. B and D are particle size
distribution histograms of Co3O4 NPs and MnOx NPs respectively.

chloroform can be attributed to the absorption of these solvents to the temperature of a catalyzed reaction is increased, it increases the colli-
surface of the NPs which inhibit aggregation of the NPs [44]. The sion frequency of the reactants by supplying them with kinetic energy
brown and black particles observed when methanol and acetonitrile which results in an increase in the collision of reactants with catalyst
were used as solvents are a result of NP aggregation [3]. While non- hence an increase in styrene conversion was observed with increase in
polar solvents gave moderate selectivity towards 2-phenylacetaldehyde temperature. A similar trend of results had been reported by Noh et al
and styrene conversion which can be attributed to the interaction of the [3]. The decrease in 2-phenylacetaldehyde selectivity can be attributed
solvent with the reactants and catalyst. to further oxidation of the product and hydrolysis to form other pro-
The increase in styrene conversion with an increase in catalyst ducts [45].
amount can be attributed to the increase in the active site of the catalyst
for the reaction to take place before they are saturated. When the

184
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

Scheme 2. The proposed reaction mechanism for styrene oxidation using metal oxide nanoparticles and immobilized nanoparticles [46].

Table 3 the metal oxide NPs, thus more active sites are available for the reaction
Comparison of metal NPs for selective oxidation of styrene. to take place. The immobilized metal oxide NPs have lower activities,
Catalyst oxidant Con. (%) Product selectivity TOF Ref.
which can be attributed to the NPs passivation [3]. Also, the change in
(h−1) binding energy due to metal to metal interaction after immobilization
STO PHA BZA others of the metal oxide NPs influenced the activities of the catalysts.
The higher styrene conversion obtained with MnOx NPs as com-
MnOx NPs TBHP 59.3 – 90.6 – 9.4 646.9 This
work
pared to Co3O4 NPs can be ascribed to the fact that Mn has a lower
Co3O4 NPs TBHP 42.3 100 – – – 338.1 This reduction potential (+1.51) for Mn3+ to Mn2+ as compared to Co
work (+1.81) for Co3+ to Co2+. Hence based on the reduction potentials, Mn
G5-OH Pd80 TBHP 64.3 67.4 – 30.2 2.4 285.1 [3] is expected to be a better catalyst for the oxidation reaction compared
G6-OH Pt160 TBHP 33.4 59.5 – 33.6 6.8 271.4 [3]
to Co. The obtained results give an insight that the product selectivity of
Pd wo =10 TBHP 47.2 73.3 – 23.8 2.9 209.3 [3]
Pt wo = 5 TBHP 27.2 70.5 – 25.0 4.0 219.5 [3] styrene oxidation depends mainly on the type of the metal used. The
Au NPs@XAD- O2 9.5 48 – 33 19 63.4 [47] difference in the structure of MnOx and Co3O4 NPs might be responsible
4-G3.0 for different product selectivity. The structure of MnOx NPs allows the
Au/MgO TBHP 62.6 54.3 16.8 10.8 18.1 297.1 [48] opening of the ring through the rearrangement of hydrogens which
(HDP)
gave rise to 2-phenyl-acetaldehyde. This can be explained by the XPS
Au/SrO (HDP) TBHP 53.0 44.8 26.0 11.7 17.4 373.8 [48]
Co/N-HCS300 O2 99 86 – – – 232.6 [7] data which showed that for Mn the 2 + oxidation state is at lower
1% Ag- H2O2 60 ˃ 99 – – – 129.2 [10] binding energy, which is the opposite to what occurs with Co [36].
METMS Hence after, immobilizing the metal oxide NPs, a mixture of products
Mn- TBHP 31 30.4 46.3 – 23.3 44.1 This
was observed because both the metal oxide NPs and the support played
Co3- work
O4_350 a role in influencing the product formed. Moreover, the interaction
Co-MnO2_350 TBHP 38.9 32.3 22.5 – 45.1 54.1 This between the metal oxide NPs and the support played a role in the
work product formed. The decrease in selectivity and conversion that was
observed after the third and fourth cycles can be attributed to the
Con. = conversion of styrene, STO = styrene oxide, PHA = 2-phenyl-acet-
leaching of the nanoparticles from the support to the solvent [3]. The
aldehyde, BZA = benzaldehyde. G5-OH and G6-OH mean fifth and sixth gen-
oxidation of styrene using metal oxide NPs and immobilized NPs on
eration hydroxyl terminated polyamidoamine dendrimers. W0 = water to sur-
mesoporous metal oxide supports was expected to follow the me-
factant ratio, HDP = homogeneous deposition precipitation,
METMS = methyltrimethoxysilane, XAD-4-G3.0 = third generation PAMAM chanism shown in Scheme 2 which was proposed by Nepak et al [46].
dendrimers on the surface of amberlite XAD-4 resin, N-HCS300 = nitrogen The two bonds between the TBHP oxygen and the nanoparticles are
doped carbon hollow spheres. formed due to enhanced electron density on the oxygen π* orbitals
caused by the positive inductive effect of the alkyl groups. The oxygen
4.2.2. Comparison of the catalytic activities of the synthesized catalysts will donate electrons to the empty d orbitals of the metal oxide nano-
The increase in styrene conversion with the increase in reaction particles. This eventually weakens the OeO bond which results in the
time for most catalysts shows that styrene oxidation requires time for formation of nanoparticle oxo species. On the metalloepoxy inter-
the reaction to take place. The higher styrene conversion observed with mediate, oxygen is transferred to the olefin double bond to form styrene
colloidal metal oxide NPs as compared to the immobilized metal oxide oxide. ß- hydride elimination at the metalloepoxy results in the for-
particles may be attributed to the high surface area to volume ratio of mation of 2-phenyl-acetaldehyde [46].

185
N. Masunga et al. Applied Catalysis A, General 559 (2018) 175–186

The catalytic activities of the colloidal MnOx and Co3O4 NPs in [6] Q. Yang, J. Zhang, L. Zhang, H. Fu, X. Zheng, M. Yuan, H. Chen, Catal. Commun. 40
styrene oxidation were compared with other reported catalysts, both (2013) 37–41.
[7] I.B. Nongwe, V. Ravat, R. Meijboom, N.J. Coville, Appl. Catal. A Gen. 466
colloidal and supported NPs systems as shown in Table 3. It was ob- (2013) 1–8.
served that MnOx and Co3O4 NPs showed higher catalytic activities as [8] S.K. Pardeshi, R.Y. Pawar, Mater. Res. Bull. 45 (2010) 609–615.
compared to other catalytic systems. The selectivity towards styrene [9] U.S. Department of Health and Human Services, Natl. Toxicol. Progr. (2011) 53.
[10] V. Purcar, D. Donescu, C. Petcu, R. Luque, D.J. Macquarrie, Appl. Catal. A Gen. 363
oxide and 2-phenyl-acetaldehyde were excellent with Co3O4 and MnOx (2009) 122–128.
NPs respectively. The immobilized metal oxide NPs on the mesoporous [11] Y. Liu, H.-J. Zhang, H.-H. Wu, X.-L. Liu, Y.-Q. Cai, Y. Lu, Chem. J. Chin. Univ. 28
metal oxide showed poor product selectivity as compared to reported (2007) 1523–1527.
[12] Y. Kaminaga, J. Schnepp, G. Peel, C.M. Kish, G. Ben-Nissan, D. Weiss, I. Orlova,
data, while its activity is comparable to literature. O. Lavie, D. Rhodes, K. Wood, D.M. Porterfield, A.J.L. Cooper, J.V. Schloss,
E. Pichersky, A. Vainstein, N. Dudareva, J. Biol. Chem. 281 (2006) 23357–23366.
5. Conclusion [13] J. Hu, K. Li, W. Li, F. Ma, Y. Guo, Appl. Catal. A Gen. 364 (2009) 211–220.
[14] S. Sharma, S. Sinha, S. Chand, Ind. Eng. Chem. Res. 51 (2012) 8806–8814.
[15] A. Rezaeifard, P. Farshid, M. Jafarpour, RSC Adv. 4 (2014) 9189–9196.
Well-defined MnOx and Co3O4 NPs were successfully synthesized [16] S. Pathan, A. Patel, Ind. Eng. Chem. Res. 52 (2013) 11913–11919.
within a narrow range of 6.2 nm and 8.4 nm respectively. The metal [17] F. Azzolina, I. Polaert, L. Estel, L.B. Pierella, Appl. Catal. A Gen. 453 (2013) 92–101.
oxide NPs were successfully immobilized on the MMOs. The colloidal [18] I. Ku, O. Rodrigues, M.A. Carvalho, I.C. Neves, A.M. Fonseca, Appl. Organomet.
Chem. 26 (2012) 44–49.
NPs showed higher activities in the catalytic oxidation of styrene as [19] R. Ghosh, X. Shen, J.C. Villegas, Y. Ding, K. Malinger, S.L. Suib, J. Phys. Chem. B
compared to the immobilized metal NPs on the mesoporous metal oxide 110 (2006) 7592–7599.
support. The catalytic activities of the compared catalysts in decreasing [20] S. Benjamin, Matthew Lane, A. Vogt, J. Victoria, K. DeRose, Burgess, J. Am. Chem.
Soc. 122 (2002) 3220–3221.
order of MnOx NPs (646.9 h−1) > Co3O4 NPs (338.1 h−1) > Co- [21] A. Askarinejad, M. Bagherzadeh, A. Morsali, Appl. Surf. Sci. 256 (2010) 6678–6682.
MnO2_350 (54.1 h−1) > Mn-Co3O4 _ 350 (44.1 h−1) ˃ MnO2_350 [22] C.-S. Li, G. Melaet, W.T. Ralston, K. An, C. Brooks, Y. Ye, Y.-S. Liu, J. Zhu, J. Guo,
(35.8 h−1) ˃ Co3O4 _ 350 (21.3 h−1) after 8 h of reaction was observed. S. Alayoglu, G.A. Somorjai, Nat. Commun. 6 (2015) 6538–6543.
[23] M.S. El-Deab, T. Ohsaka, Angew. Chem. Int. Ed. Engl. 45 (2006) 5963–59676.
MnOx NPs showed high selectivity towards 2-phenyl-acetaldehyde [24] J. Hu, K. Li, W. Li, F. Ma, Y. Guo, Appl. Catal. A Gen. 364 (2009) 211–220.
while the Co3O4 NPs showed 100% selectivity towards styrene oxide. [25] Q. Tang, Y. Wang, J. Liang, P. Wang, Q. Zhang, H. Wan, Chem. Commun. (2004)
The immobilized metal oxide NPs did not show any selectivity towards 440–441.
[26] C. Xu, M. Lu, B. Yan, Y. Zhan, P. Balaya, L. Lu, J.Y. Lee, ChemSusChem 9 (2016)
a specific product. After 8 h of reaction, the TEM images of the NPs 3067–3073.
showed their stability. Therefore, the catalyst in this study can be re- [27] L. Wu, Q. Li, C.H. Wu, H. Zhu, A. Mendoza-Garcia, B. Shen, J. Guo, S. Sun, J. Am.
cycled and used for further reactions. Also, the recycle test showed that Chem. Soc. 137 (2015) 7071–7074.
[28] Y.-X. Li, Z.-Y. Wei, L. Liu, M.-L. Gao, Z.-B. Han, Inorg. Chem. Commun. 88 (2018)
the immobilized metal oxide NPs are stable, and the product selectivity
47–50.
and styrene conversion did not change much. Therefore, styrene oxi- [29] F. Zhang, X. Zhao, C. Feng, B. Li, T. Chen, W. Lu, X. Lei, S. Xu, J. Catal. (2011)
dation using these metal oxide NPs can contribute to the scientific 232–237.
knowledge and development of Co3O4 NPs and MnOx NPs based cata- [30] J. Park, K. An, Y. Hwang, J.-G. Park, H.-J. Noh, J.-Y. Kim, J.-H. Park, N.-M. Hwang,
T. Hyeon, Nat. Mater. 3 (2004) 891–895.
lytic systems in the field of oxidation catalysis. [31] W. Huang, J.N. Kuhn, C.K. Tsung, Y. Zhang, S.E. Habas, P. Yang, G.A. Somorjai,
Nano Lett. 8 (2008) 2027–2034.
Acknowledgments [32] D. McNulty, H. Geaney, C. O’Dwyer, Sci. Rep. 7 (2017) 42263.
[33] J.F. Marco, J.R. Gancedo, M. Gracia, J.L. Gautier, E. Ríos, F.J. Berry, J. Solid State
Chem. 153 (2000) 74–81.
Authors are grateful for the financial support we received from the [34] T. Choudhury, S.O. Saied, J.L. Sullivan, A.M. Abbot, J. Phys. D Appl. Phys. 22
University of Johannesburg and the National Research Foundation of (1989) 1185–1195.
[35] A. Pendashteh, J. Palma, M. Anderson, R. Marcilla, RSC Adv. 6 (2016)
South Africa (Grant specific numbers 5386 and 93205). We would like 28970–28980.
also to thank Mr. D. Hariss and Dr. R. Meyer from Shimadzu South [36] T.H. Dolla, K. Pruessner, D.G. Billing, C. Sheppard, A. Prinsloo, E. Carleschi,
Africa for the equipment we used to conduct this project. B. Doyle, P. Ndungu, J. Alloys Compd. 742 (2018) 78–89.
[37] B.M. Mogudi, P. Ncube, R. Meijboom, Appl. Catal. B Environ. 198 (2016) 74–82.
[38] A.S. Poyraz, C.-H. Kuo, S. Biswas, C.K. King’ondu, S.L. Suib, Nat. Commun. 4 (2013)
Appendix A. Supplementary data 2952–2962.
[39] N. Masunga, G.S. Tito, R. Meijboom, Appl. Catal. A Gen. 552 (2018) 154–167.
[40] X. Tang, J. Chen, Y. Li, Y. Li, Y. Xu, W. Shen, Chem. Eng. J. 118 (2006) 119–125.
Supplementary material related to this article can be found, in the
[41] X. Tang, J. Chen, X. Huang, Y. Xu, W. Shen, Appl. Catal. B Environ. 81 (2008)
online version, at doi:https://doi.org/10.1016/j.apcata.2018.04.027. 115–121.
[42] A. Wang, H. Jing, Dalton Trans. 43 (2014) 1011–1018.
References [43] B. Tyagi, B. Shaik, H.C. Bajaj, Appl. Catal. A Gen. 383 (2010) 161–168.
[44] M.R. Johan, K. Si-wen, N. Hawari, N. Azri, K. Aznan, Int. J. Electrochem. Sci. 7
(2012) 4942–4950.
[1] N.T. Thao, N.D. Trung, D. Van Long, Catal. Lett. 146 (2016) 918–928. [45] N.T. Thao, H.H. Trung, Catal. Commun. 45 (2014) 153–157.
[2] M. Nemanashi, R. Meijboom, Catal. Lett. 143 (2013) 324–332. [46] D. Nepak, D. Srinivas, Appl. Catal. A Gen. 523 (2016) 61–72.
[3] J.-H. Noh, R. Patala, R. Meijboom, Appl. Catal. A Gen. 514 (2016) 253–266. [47] A.S. Sharma, D. Shah, H. Kaur, RSC Adv. 5 (2015) 42935–42941.
[4] F. Zhang, X. Zhao, C. Feng, B. Li, T. Chen, W. Lu, X. Lei, S. Xu, ACS Catal. 1 (2011) [48] N. Patil, B. Uphade, P. Jana, S. Bharagava, V. Choudhary, J. Catal. 223 (2004)
232–237. 236–239.
[5] N.T. Thao, L. Thi, K. Huyen, Chem. Eng. J. 279 (2015) 840–850.

186