Manganese nanocatalyst and N-hydroxyphthalimide as an efficient catalytic system for selective oxidation of ethylbenzene, cyclohexene and oximes under aerobic condition

Article abstract of DOI:10.1016/j.molcata.2013.10.023

The catalytic activity of the nano-manganese-catalyst in the aerobic oxidation of ethylbenzene, cyclohexene and oximes were evaluated without the need of any reducing agent, using N-hydroxyphthalimide (2-Hydroxy-1H-isoindole- 1,3-dione; NHPI) under oxygen pressure. The oxidation of ethylbenzene and cyclohexene to acetophenone and 2-cyclohexene-1-one using N-hydroxyphthalimide (2-Hydroxy-1H-isoindole-1,3-dione; NHPI) under oxygen atmosphere in the presence of a SiO₂/Al₂O₃-supported manganese catalyst occurs with conversions of 95 and 99% and selectivities of 99%. The supported manganese is also a suitable and efficient catalyst for the oxidative deprotection of oximes to the corresponding carbonyl compounds. A mild and efficient method has been optimized for manganese catalysts by considering the effect of various parameters such as the reaction time and the amount of catalyst, the temperature and the reusability of the catalyst after several runs without modification. Moreover, some possible mechanisms for the oxidation of ethylbenzene, cyclohexene and oximes have been proposed.

Full text of DOI:10.1016/j.molcata.2013.10.023

Accepted Manuscript
Title: Manganese catalyst and N-hydroxyphthalimide as an
efficient catalytic system for selective oxidation of
ethylbenzene, cyclohexene and oximes under aerobic
condition
Author: D. Habibi A.R. Faraji M. Arshadi H. Veisi A. Gil
PII:
S1381-1169(13)00406-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2013.10.023
MOLCAA 8916
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
26-7-2013
24-10-2013
26-10-2013
Please cite this article as: D. Habibi, A.R. Faraji, M. Arshadi, H. Veisi, A.
Gil, Manganese catalyst and N-hydroxyphthalimide as an efficient catalytic
system for selective oxidation of ethylbenzene, cyclohexene and oximes
under aerobic condition, Journal of Molecular Catalysis A: Chemical (2013),
http://dx.doi.org/10.1016/j.molcata.2013.10.023
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

*Graphical Abstract (for review)
Efficient catalytic systems based on manganese for oxidation of
ethylbenzene, cyclohexene and oximes in the presence of N-
hydroxyphthalimide
1
Page 1 of 54

*Highlights (for review)
Highlights
Mn supported on SiO2-Al2O3 used as catalysts using N-hydroxyphthalimide.
Mn catalysts were studied in the oxidation of ethylbenzene to acetophenone.
Mn catalysts were studied in the oxidation of cyclohexene to 2-cyclohexene-1-one.
Conversions reached to 95 and 99%, and selectivities of 99%.
Mn catalysts for the oxidative deprotection of oximes to carbonyl compounds.
Page 2 of 54

*Manuscript
Manganese catalyst and N-hydroxyphthalimide as an efficient
catalytic system for selective oxidation of ethylbenzene, cyclohexene
and oximes under aerobic condition
D. Habibi1*a, A. R. Faraji*a, M. Arshadib, H. Veisic, A. Gild
a Department of Organic Chemistry, Faculty of Chemistry, Bu-Ali Sina University, Hamedan
6517838683, Iran
b Department of Science, Fasa Branch, Islamic Azad University, PO Box 364, Fasa 7461713591,
Fars, Iran
c Department of Chemistry, Payame Noor University, 19395-4697 Tehran, Iran
d Department of Applied Chemistry, Los Acebos Building, Public University of Navarra,
Campus of Arrosadia, E-31006, Pamplona, Spain
Abstract
The catalytic activity of the nano-manganese-catalyst in the aerobic oxidation of ethylbenzene,
cyclohexene and oximes were evaluated without the need of any reducing agent, using N-
hydroxyphthalimide (2-Hydroxy-1H-isoindole-1,3-dione;NHPI) under oxygen pressure. The
oxidation of ethylbenzene and cyclohexene to acetophenone and 2-cyclohexene-1-one using N-
hydroxyphthalimide (2-Hydroxy-1H-isoindole-1,3-dione;NHPI) under oxygen atmosphere in the
presence of a SiO2/Al2O3-supported manganese catalyst occurs with conversions of 95 and 99%
and selectivities of 99%. The supported manganese is also a suitable and efficient catalyst for the
oxidative deprotection of oximes to the corresponding carbonyl compounds. A mild and efficient
method has been optimized for manganese catalysts by considering the effect of various
parameters such as the reaction time and the amount of catalyst, the temperature and the
1 *Corresponding author. Tel.: +98 811 8282807; fax: +98 811 8380709.
E-mail address: davood.habibi@gmail.com (D. Habibi) and alireza_ch57@yahoo.com.
1
Page 3 of 54

reusability of the catalyst after several runs without modification. Moreover, some possible
mechanisms for the oxidation of ethylbenzene, cyclohexene and oximes have been proposed.
Keywords: Manganese supported catalyst; Oxidation; Ethylbenzene; Cyclohexene; Oxime.
1. Introduction
Selective oxyfunctionalization of substrates to give oxygen containing compounds is a
fundamental reaction in organic chemistry for fundamental research and industrial
manufacturing. Direct oxidation of the abundant and cheapness hydrocarbons containing
benzylic and allylic carbon-hydrogen bonds into corresponding ketones are of major important
transformations in organic synthesis, since the oxidation products are essential intermediates for
the generation of the useful primary, special chemicals and high economic value fine chemicals,
agrochemicals, pharmaceutical and perfumes industry [1]. Selective oxidation of organic
substrates with eco-friendly and readily available oxidants instead of hazardous oxidant such as
peroxides, permanganate sodium hypochlorite, iodosobenzene and dichromate have received
recently considerable attention [2,3]. In the last few years, due to apparent advantages of using
O2 as oxidant, considerable attempt has been invested to improve efficient catalytic system for
the aerobic oxidation of organic compounds. The dominant position of O2 as the oxidant for bulk
chemical oxygenations is due to the fact that it is the only economically and environmentally
feasible oxidant for large scale processing. The success of these oxy-functionalizations always
depends largely on the development of catalysts to promote productivity, the rate of reaction, as
the selectivity of products. The heterogenisation of metal complexes with catalytic activity
within inorganic matrixes (such as SiO2, clay, MCM-41, SBA-15, mesoporous materials, zeolite,
2
Page 4 of 54

etc) is a field that has progressed in last decades due to the combination of high reactivity and
chemical selectivity with facile separation of the catalyst from the reaction mixture. Due to its
properties, particular attention has been paid to reusability, thermal stability and low cost of the
catalysts.
In recent years, there has been an increased interest in the aerobic oxidation of allylic and
benzylic carbon-hydrogen bonds using transition metal Schiff base complexes immobilized on
several solid supports. Schiff base complexes are excellent chemical and thermal stability, easily
accessibility and synthesized [4,5]. These wide ranges of applications of metal complexes of
Schiff base in homogeneous and heterogeneous catalysis have generated a great deal of interest
in the synthesis of new metal complexes. Investigation related to the new solid oxidation
catalysts is one of the most significant current topics, connected with both industrial and
fundamental researches, because the traditional techniques involve the use of large amounts of
highly toxic oxidation state of Os, Mn, and Co reagents [6]. In fact, as the traditional metal
compounds used in this process are in catalytic amounts, the reaction conditions are harsh; the
product selectivity is low and often corrosive promoters are used along with the traditional
catalyst. However, the separation of the catalyst from the reaction mixture is very difficult, the
catalyst cannot be reused, low conversion and selectivity and also a lot of waste is produced
[1,7,8]. These disadvantages were minimized if homogeneous complexes were immobilized on
the surface of solid supports. Much effort has been made to activate molecular oxygen by several
catalysts or transition metal complexes, in particular the Schiff base ones, for the catalytic
oxidation of organic substrate under mild condition. In past decade, N-hydroxyphtalimide
(NHPI) has considerable interest as a novel organo-catalyst and co-catalyst for a wide variety of
free radical out oxidant [9]. With this oxidation system, in oxidation of organic substrates
3
Page 5 of 54

containing reactive carbon-hydrogen bonds, e.g. alkylbenzene [10] and alkyl aromatic
compounds containing benzylic functions [11-13], oxidation reactions catalyzed by NHPI in the
presence of transition metal as a co-catalyst [14-22] have been studied. In the proposed
mechanisms, the phthalimide-N-oxyl radical (PINO) is formed by abstracting H atoms from the
NHPI, then PINO can in turn abstract H atoms from the substrate [23-38]. In the present
investigation, we report a simple procedure for preparation of a type Mn catalyst immobilized
the Schiff base ligands (Scheme 1), and apply this catalyst with homogeneous NHPI for direct
aerobic oxidation of ethylbenzene and cyclohexene. When concerning their catalytic
performances for the oxidation of ethylbenzene, cyclohexene and oximes under mild conditions
with excellent conversion and high selectivity, the catalyst show promising activities for the
reactions as expected.
2. Experimental
2.1. Materials and characterization techniques
All reagents were purchased from the Merck and Fluka chemical companies. Reagents
were used without extra purification, but solvents were purified with standard methods. All
compounds were known and identified by comparison of their physical and spectroscopic data
with those of authentic samples. Inductively coupled plasma (ICP) measurements for manganese
content evaluation were performed using a Perkin-Elmer ICP/6500. Infrared was collected on
KBr pellets using a JASCO FT/IR (680 plus) spectrometer. Diffuse reflectance spectra were
registered on a JASCO-550 UV-Vis spectrophotometer that was equipped with a Diffuse
reflectance attachment in which BaSO4 was as the reference. XPS (small area X-ray
photoelectron spectroscopy) data were recorded with the PHI-5702 Multi-Technique System.
4
Page 6 of 54

The thermal analysis was recorded on a Perkin-Elmer TG-DTA 6300 instrument at a heating rate
of 15◦C.min-1 and for elemental analysis a CHN-Rapid Heraeus elemental analyzer (Wellesley
MA) was used. Before carrying out the nitrogen (99.999 %) adsorption experiments, the sample
was outgassed at 393 K for 14 h, then the experiment have been carried out at 77 K using a
volumetric apparatus (Quantachrome NOVA automated gas sorption analyzer). The specific
surface areas were calculated using the BET method. The images of scanning electron
micrograph (SEM) and transmission electron microscopy (TEM) were taken using a Philips 501
microscope and a Tecnai F30TEM operating at 300 kV, respectively. In addition, energy
dispersive X-ray analysis was conducted on each sample. Size distribution was measured in
order to number of nanoparticles in reverse microemulsion by Zetasizer Nano-ZS-90 (ZEN 3600,
MALVERN instruments). The reaction products of aerobic oxidation were determined and
analyzed by a HP 6890/5973 GC/MS instrument and analyzed by a Shimadzu GC-16A gas
◦
chromatograph (GL-16A gas chromatograph with a 5m×3mm OV-17 column, 60–220 C (10
◦C/min), Inj. 230 ◦C, Det. 240 ◦C).The products were determined by comparing with the standard
mass spectrometry of organic compounds and fragmentation pattern.
2.2. Preparation of organometallic-SiO2/Al2O3
SiO2/Al2O3 was used as the support prepared by the sol-gel method. First, some defined
value of Tetra-Ethyl-Ortho-Silicate (TEOS) and Aluminium butyrate in n-butanol were
dissolved. The final mixture has a standard mole ratio of Si(-OC2H5)4–Al(-OC4H9)3–C4H9OH in
the range of 15(Si(-OC2H5)4) – 10(Al(-OC4H9)3) –15 (C4H9OH). The obtain solution was heated
up to 70 ºC and the components were entirely mixed. Then it was cold down to the room
5
Page 7 of 54

temperature. 2,4-Pentandione (H-acac) as the complacing agent was added to it. In this way, the
apparent solution was obtained. After that by adding deionized water (12.0 mol H2O/mol
alkoxide) to the obtained solution, the solution was hydrolyzed and the apparent gel was
produced. Then for removing the solvent and water from the gel, it was dehydrated up to 100 ºC.
Also for eliminating of organic compounds, it was calcined at 550 ºC for 5 h. So, the synthesized
solid SiO2/Al2O3 was being used as the support in synthesis of the catalyst [34]. At first, 3.5 g of
SiO2/Al2O3 was activated at 500 ºC for 5 h under air and then was refluxed with 4.3 cm3 of
trimethoxysilylpropylamine (APTMS) in dry toluene (50 cm3) for 24 h. The solid achieved
during this process was filtered and washed off with dry methanol at 100 ºC under vacuum for 5
h. Then, bipyridylketone (BPK) was added to a suspended solution of SiO2/Al2O3-APTMS in dry
methanol. To synthesize of SiO2/Al2O3-APTMS-BPK-Mn, 2.0 g SiO2/Al2O3-APTMS-BPK was
suspended in 50 cm3 of ethanol in a round bottom flask followed by adding of 3.0 mmol
Mn(OAc)2·4H2O. The mixture was refluxed during 24 h under magnetically stirring.
2.3. General procedure for oxidation of ethylbenzene and cyclohexene
A suspension of the heterogeneous catalyst (0.1 g), solvent (5 cm3 of acetic acid/
benzonitrile), 2 mmol substrate and N-hydroxyphthalimide (NHPI) (15 mol %) were mixed in a
three necked round bottom flask which was fitted with a equipped water condenser through a
balloon filled with O2. The liquid phase oxidation reactions were carried out at desired
6
Page 8 of 54

temperature with vigorous stirring. After filtration and washing with solvent, the type and
quantity of the resulting products from oxidation were determined by GC-MS instrument.
Besides, a comparative experiment with various conditions was carried out.
2.4. General procedure for oxidation of oximes
In this procedure, a suspension of the heterogeneous catalyst (0.07 g), solvent (5 cm3 of
toluene), oxime (2 mmol) and benzaldehyde (6 mol) were mixed in a three necked round bottom
flask which was fitted with a equipped water condenser with a balloon filled of O2. The liquid
phase oxidation reactions were carried out at desired temperature with vigorous stirring. After
completing the reaction, the catalyst has been separated through filtering and the products and
the amounts of them have been quantified by GC-MS and GC. Finally, a suitable reaction
condition has been optimized.
3. Results and discussion
3.1. Synthesis of manganese-supported catalyst
Using BPK and Mn(OAc)2·4H2O through the covalently immobilization, the
heterogeneous manganese catalyst was synthesized on the SiO2/Al2O3 mixed oxides, as
illustrated in Scheme 1. The formation of manganese catalyst on the SiO2/Al2O3 was verified
using CHN analysis, FT-IR, UV–vis, Scanning Electron Microscope (SEM), Transmission
7
Page 9 of 54

Electron Microscope (TEM), X–ray photoelectron spectroscopy (XPS), Energy-dispersive X-ray
spectroscopy (EDX), Dynamic light scattering (DLS), Inductively coupled plasma (ICP),TGA-
DTA and nitrogen adsorption. The elemental analysis report along with summary of measured
and calculated values for C, H and N is presented in Table 1. Analysis of the results indicated the
fact that the CHN content increased with the organic chain size because of immobilization of
organic groups in the inorganic matrix. The loading of manganese in the heterogeneous
manganese catalyst was characterized by elemental analyses. The final metal content was 0.35
mmol/g, indicating that 74.4 % of the immobilized ligands were complexed with manganese
ions.
Analysis of the specific surface area was performed by standard Brunauer-Emmett-Teller
technique (BET) for analysis of gas adsorption using nitrogen. Nitrogen adsorption experiments
were performed using 1 g of the SiO2/Al2O3. The specific surface area, pore size and pore
volume of the modified support were significantly decreased compared to the parent SiO2/Al2O3.
This behaviour indicates the decrease in interaction between molecular nitrogen as an adsorbate
and the parent SiO2/Al2O3 surface after modification with organic groups. Furthermore, the
isotherm for support is type I, which is characteristic for microporous materials. The micropores
have a size of 0.60 nm and a rather broad pore size distribution (see Table 1).
The FT-IR spectra of SiO2/Al2O3-APTMS, SiO2/Al2O3-APTMS-BPK, and SiO2/Al2O3-
APTMS-BPK-Mna are shown in Figure 1. The strong absorption bands related to Si-O-Si
8
Page 10 of 54

stretching vibrations are observed in the spectrum of SiO2/Al2O3 at 798 and 1010-1290 cm-1. The
FT-IR spectrum of SiO2/Al2O3-APTMS shows several signals originating from amino propyl
groups, which are related to C-H stretching modes of the propyl, appeared in the area of 1450-
1560 cm-1 and 2860-2935 cm-1. By observing the FTIR results it could be concluded that the
SiO2/Al2O3 was successfully modified by amine spacer groups. The N-H deformation peak at
1540-1560 cm-1 confirms the successful functionalization of the Si/Al mixed oxide with 3-
APTMS. The C=N imine vibration signal was observed at 1630 cm-1, which shows the
condensation reaction between BPK with organo-functionalized SiO2/Al2O3 (see Figure 1-B).
The peaks in the 3020-3066 cm-1 ranges and the peaks at 1434-1437 cm-1 are attributed to the C-
H stretching vibrations of pyridine groups and C=C stretching vibration of pyridine groups.
Furthermore, after complexing of manganese with immobilized BPK over modified SiO2/Al2O3,
the weak absorption peak at 413 cm-1 was appeared that is attributed to the Mn-N bands [39,40].
The UV-Vis spectrum of the SiO2/Al2O3 had a side-band adsorption near 249 nm. In the
SiO2/Al2O3-APTMS-BPK transitions π→π* and n→π* of the ligands caused strong adsorption
in the 255-320 nm (see Figure 2). After the reaction of SiO2/Al2O3-APTMS-BPK with
Mn(OAc)2·4H2O, the color of the reaction mixture was changed from yellow to deep brown.
Consequently, UV-visible spectra of the immobilized Mn(III) species indicated the appearance
of several new metal d–d migration bands around 490 nm upon complexation.
9
Page 11 of 54

A selected image of scanning electron microscope (SEM) and transmition electron
microscope (TEM) of catalyst is shown in Figure 3, from which it can be seen that the
nanoparticles appearance and size of them were similar, demonstrating that the particles of SiO2-
Al2O3 have good mechanical stability and they have not been destroyed during the whole
modification. However, the average particle size in reverse micro emulsion solution of the
catalyst was around 31 nm. According to the small nanoparticle size and ligand capping as an
obstacle in agglomeration, the Mn-catalyst could be used as a suitable catalyst for oxidation of
several substrates such as ethylbenzene, and cyclohexene. The energy dispersive spectrum (EDS)
of Mn-catalyst was shown in Figure 3. In the EDS spectrum of the cobalt catalyst, signals related
to Si, Al, O and Mn were observed. The existence of Mn signal in the spectrum resulted from the
Mn complexation with active sites of the organic functional groups (-C=N) that increased the
catalytic activity of the synthesized catalyst in comparison with the unmodified support.
Thermal stability of the synthesize materials were studied with TGA (thermo gravimetric
analysis) and DTG (derivative thermogravimetric) in Ar atmospheres in the range of 25-1200 °C
with a rate of heating of 10 °C/min (see Figure 4) [32]. Thermal analysis of nanoscale
SiO2/Al2O3 showed two stages of weight loss, stage I (25-115 ºC) corresponds to the desorption
of physically held water (2.6 wt.%), on the surface of SiO2/Al2O3, and Stage II (120-1200 ºC) is
due to the condensation of Al-OH and Si-OH (13.4 wt.%), present in the structure of support (for
example; 2Si-OH → Si-O-Si + H2O). However, Mn catalyst showed four distinct steps: at 25-89,
10
Page 12 of 54

89-291, 291-556 and 556-1200 ºC (see Figure 4). Stages II and III (89-556 ºC) are due to
desorption of water molecules from the pores of SiO2/Al2O3 and also decomposition of the
immobilized organic functional groups (-pr-NH2). The additional weight losses in stage IV could
be observed due to the several phenomena; the calcination of coke, the loss of the new generated
Al-OH and Si-OH groups and decomposition of the immobilized Mn complex groups. From
thermogram analysis of samples, it could be concluded that thermal stability of SiO2/Al2O3 in
Mn catalyst improved rather than unmodified SiO2/Al2O3 and the weight losses of the modified
SiO2/Al2O3 with Mn complex groups in the temperature range of dehydration are less than that
for unmodified SiO2/Al2O3.
The XP spectroscopic analysis of the heterogeneous catalyst was used to study for the
characterization of oxidation state of the active sites of immobilized manganese complexes on
the surface of SiO2/Al2O3. This technique has previously been shown to provide valuable
information regarding the chemical state of the catalytically active sites in various catalysts [41-
44]. The XPS spectrum of Mn catalyst produced a Mn 2p3/2 and a Mn 2p1/2 peak at 643.01 eV
and 654.61 eV, and is depicted in Figure 5. These results indicated that most of the Mn ions on
the surface of the heterogeneous catalyst were in +3 oxidation state (due to the binding energy
(BE) 643.01 eV which are attributed to Mn(III) state of manganese), and they are in accordance
with earlier literature data [41-43].
11
Page 13 of 54

3. 2. Oxidation of ethylbenzene
The oxidation of ethylbenzene with oxygen by the new catalytic system using a
heterogeneous transition metal in the presence of NHPI was investigated. The reaction products
were identified by comparison with authentic products and GC-MS. For ethylbenzene oxidation,
acetophenone (AcPO) and -phenylethanol (PEA) were determiated. The effects of various
operating parameters, such as the amount of the catalyst to NHPI, the temperature, and the
reaction time, on the performance of the heterogeneous manganese-catalyst was investigated in
the oxidation of ethylbenzene by O2 in the presence of the acetic acid as solvent. In fact, due to
its polar nature, NHPI has a low solubility in ethylbenzene at the mild temperature employed in
our protocol. Thus, the use of acetic acid is mandatory in order to dissolve the quantities of NHPI
necessary for observing a good catalytic activity. Acetophenone and α-phenyl ethanol (Scheme
2) were the major reaction products obtained. The GC analysis did not show any oxidation
products of the aromatic ring in effluents.
Several solvents were examined in the present aerobic oxidation. The aerobic oxidations of
ethylbenzene must be carried out in an appropriate solvent such as acetic acid or acetonitrile due
to the lower solubility of NHPI in nonpolar solvents such as hydrocarbons. Among the various
solvents, acetic acid was found to be the optimum solvent [35-38]. In order to assess the effect of
catalyst to oxidant molar ratio on the catalytic activity and selectivity in acetic acid under
oxygen, the reactions were studied at various amounts of catalyst:NHPI (see Table 2). It was
12
Page 14 of 54

observed that catalytic activity and selectivity is strongly affected by increasing amount of
catalyst to NHPI. The most promising observation is that acetophenone was found to be the
major product in most of the catalytic reactions. The temperature and time dependences of
ethylbenzene oxidation, employing Mn(III)/NHPI as the catalytic system, are presented in Table
2. Ethylbenzene was oxidized at 25 °C and the conversion was 19%. When the temperature was
increased to 100 °C, the conversion of ethylbenzene was significantly improved to 53%. The
results indicated that the temperature is another key factor in the selective oxidation of
ethylbenzene. Therefore, the appropriate temperature for oxygenation catalyzed by
Mn(III)/NHPI system should be 100–120 °C. In all EB oxidation reaction conditions, conversion
and selectivity to acetophenone increased with reaction time.
Table 2 shows the conversion and selectivity do rise with rising concentration of
manganese catalyst to EB and NHPI, when the reaction temperature is fixed, that is, due to the
presence of excess of active sites on the catalyst. The optimized conditions were found to be as
follows: ethyl benzene (2 mmol), catalyst (0.1 g), NHPI (15 % of substrate) and acetic acid (5
ml) under pressure of oxygen at 120 °C. The oxidation of ethylbenzene in the absence of catalyst
gives low conversion and selectivity to acetophenone under oxygen pressure at 25 °C (see Table
2, entry 1).
NHPI has been demonstrated to be a free radical oxidation catalyst [11,12]. In contrast to
transition metals, NHPI does not accelerate the hydroperoxide decomposition reaction; therefore,
13
Page 15 of 54

the catalytic activity of NHPI results from phthalimido-N-oxyl radical (PINO) formation in the
propagation step of the oxidation process. PINO radical is highly electrophilic species and can
efficiently promote hydrocarbon oxyfunctionalization with molecular oxygen.
In fact, transition metals, such as cobalt and manganese, are typically used as catalysts in
these oxidation processes, that is, their catalytic effect lowers the activation energy of the
decomposition reaction of the hydroperoxide. A plausible reaction pathway for the aerobic
oxidation of ethylbenzene by the combined use of NHPI and manganese catalyst is proposed in
Scheme 3, at first step, NHPI was easily converted into reactive PINO radicals under reaction
conditions, and promoted radical transfer and chain rupture the benzylic hydrogen atom of
ethylbenzene. Futheremore, Mn(III) ions supported on the catalyst were able to accelerate NHPI
to form PINO radicals [45-48]. Furthermore, for demonstrating the role of the Mn-catalyst
pyridine molecule was added into the system which resulted in decreasing the catalytic activity
of manganese complex quickly, in fact, the nitrogen of pyridine molecule easily occupies the
active sites of the catalyst. In the second step, NHPI in situ could be converted to PINO by the
reaction of NHPI with the catalyst-oxygen complex that would be the most important step in the
present oxidation [38]. The BDE (Bond Dissociation Energy) value of the O-H bond for NHPI is
> 86 kcal/mol by means of ESR spectroscopy [49,50]. The BDE of the benzylic C-H bond in
ethylbenzene is estimated to be about 87 kcal/mol. This suggests that PINO could abstract the
benzylic hydrogen atom of ethylbenzene; therefore, at third step, PINO abstracts the hydrogen of
14
Page 16 of 54

a C-H bond in ethylbenzene to produce carbon-centered radicals, and then propagates the
autoxidation. However capturing the obtaining ethylbenzyl radicals by dioxygen molecules
provides alkylperoxy radicals, which are eventually converted into oxygenated compounds
through 1-phenylethyl hydroperoxide (PEHP) (step 3). This intermediate can produce AcPO and
PEA in two paths (steps 4 and 6), at step 5, alkylproxy radical can abstract hydrogen from NHPI
and at step 6, two alkylproxy radicals combine to form a tetroxide intermediate, which further
decomposes through a 6-membered ring transition state rearrange to a molecule of oxygen, an
acethophenone and phenylalcohol; well-known as Russell termination. It is expected from this
mechanism that the ratio of these products are 1.0. During the first 2 h of the reaction, ketone and
alcohol formation takes place only via Russell termination but at higher conversions it is clear
that other mechanisms play an important role [11,12,38]. Mn(II) were reoxidized to ones with
higher valences during the course of the oxidation. The oxidants for low-valence Mn species
were suggested to be the intermediate alkyl hydroperoxide and molecular oxygen [49]. The
concentration of alkyl hydroperoxide decreased quickly, because it was catalytically decomposed
by the manganese complex. It has been reported that the acidity of catalyst plays an important
role for dehydration of PEHP to AcPO [51]. Therefore, the selectivity of hydroperoxide in acetic
acid decreased after the increase in reaction time or temperature. Therefore in the next step, the
PEHP was catalytically decomposed by the Mn(II) complex, so, the selectivity of PEHP
decreases after the increase of reaction time. The selectivity of AcPO increases with the catalyst
15
Page 17 of 54

amount (entry 5 and 9, table 2). This observation indicated that as the amount of the Mn complex
increases, the decomposition of PEHP becomes more rapid [35,38].
In relation with the above discussion, it could be concluded that both reduction and
reoxidation of manganese species of the catalyst occurred together to construct the redox cycle in
the oxidation. However, in the presence of acetic acid, the rate of this reaction increased and the
selectivity of PEHP decreased, that is, the proton of solvent can prompt decomposition of PEHP,
and eventually this step becomes more rapid [35,38]. The PEHP in the next steps further
converts to acetophenone and 1-phenylethanol (steps 7 and 8).
The high selectivity to acetophenone and the absence of 2-phenylacetaldehyde can be
rationalized from the following discussion. The methyl group carries more number of hydrogen
than the methylene group that was not oxidized. However, the methyl group might be rotating
more rapidly than the methylene group and hence its hydrogen was not abstracted by activated
oxygen as shown in Scheme 3. The existence of a barrier to rotation about Csp -Csp3 was already
2
reported [52,53]. Therefore, the hydrogen of methylene group was more readily available for
oxidation than that of methyl group. The distant chemisorbed oxygen abstracts the hydrogen
from the methylene group of ethylbenzene and fashions phenyl ethyl radical and metal
hydroperoxide. The free radical rapidly reacts with metal hydroperoxide to form PEA. The local
magnetic field of active sites of heterogeneous catalyst (Mn ions) could be thought to retain the
paramagnetic phenyl ethyl radical until it is hydroxylated. The obtaining PEA is rapidly acted
16
Page 18 of 54

upon by the chemisorbed oxygen on the metal active site or PINO to form the product radical as
shown in the reaction (Scheme 3). The oxidation of PEA might be more rapid than ethylbenzene,
as the rotation of -CHOH group is slower than the -CH2- group. Furthermore, the polar nature of
transition state model between PINO and benzylic hydrogen confirmed the high reactivity of
PEA compared to ethylbenzene. In fact, the benzylic OH-group of PEA stabilizes the partial
positive charge on the benzylic carbon atom by resonance, and therefore favors further oxidation
to AcPO. Consequently, the hydrogen atom which reacted with active sites of Mn-supported
catalyst was ejected by radical in order to form H2O and AcPO.
Besides coordination to a Schiff base ligand, the Mn ions may also interact with the surface
Si–OH and Al–OH groups of the SiO2/Al2O3. Thus, the materials of blank SiO2/Al2O3 that
adsorbed Mn acetate were also used for oxidation experiment. It turned out that the conversion of
ethylbenzene was almost 8.5% (similar to the result of blank experiments, Table 2, entry 1).
Therefore, in the oxidation of ethylbenzene with NHPI and oxygen under acetic acid, the
manganese ions absorbed on the surface of the support did not contribute in the oxidation
reaction. The above observations suggest that the oxidation occurs due to the catalytic nature of
the chemically immobilized of Mn catalysts onto the Schiff base ligands.
To assess the reusability, the catalyst was separated by filtration after the first run, washed
with ether and dried at 120 °C under vacuum and then used for the next runs under the same
conditions (see Fig. 6). No significant loss of activity and selectivity was observed, confirming
17
Page 19 of 54

that the Mn-supported catalyst has high stability during the oxidation process. To further proof
that the reaction was catalyzed by the manganese sample, we added extra ethylbenzene to the
filtrate after the removal of the catalyst and no more products were produced under the same
conditions. Based on the above, we reasoned that manganese catalyst and NHPI could couple to
be an effective catalytic system for ethylbenzene oxidation.
3.3. Oxidation of cyclohexene
To extend the present method, the oxidation catalytic activity of the manganese catalyst
has been examined using NHPI and oxygen under benzonitrile as solvent at several temperatures.
After the reaction, the products were identified and quantified by GC-MS. The products of the
reaction were cyclohexene oxide, 2-cyclohexene-1-one and 2-cyclohexene-1-ol (see Scheme 4).
Several results of the cyclohexene conversion and the selectivity of the products are
summarized in Table 3. The oxidation of CH using NHPI proceeded even at 25 ºC to give 2-
cyclohexene-1-one in higher selectivity after 24 h (76.6 %). As the reaction time were raised
from 2 to 10 h the CH conversion and the selectivity to 2-cyclohexene-1-one increased from 66
to 84 and 93 to 99 %, respectively, while the selectivity to cyclohexene oxide and 2-
cyclohexene-1-ol were decreased at 120 ºC. In the oxidation of cyclohexene, all the reaction
conditions showed excellent selectivity for the ketone with only trace amounts (in total < 1.0 %)
of by-products after 16 h at 100 ºC.
18
Page 20 of 54

The oxidation of CH in the absence of catalyst gives low conversion and selectivity to 2-
cyclohexene-1-one at 25 °C (entry 1). Furthermore, no reaction was observed in the absence of
NHPI because no radical formations from manganese centers occur under the conditions selected
in this work [12]. Reaction carried out to some extent in the absence of the catalyst, but the
activity is lower (4 %) than when this compound is present. NHPI oxidizes CH slowly, and
selectivity to 2-cyclohexene-1-one was not high after 24 h (58.9 %). This result is in accord to
the previously reported reaction mechanisms [12,35], where NHPI are responsible for radical
formation and the heterogeneous manganese is the co-catalyst that enhance the activity of the
NHPI. Therefore, these could be demonstrated again from the results that under the mentioned
reaction conditions, the allylic hydrogen is more reactive than the C=C double bond. However,
the abstraction of hydrogen from the allylic carbon leads to allylic radical, which requires lower
activation energy than the reaction at double bond.
Reusability of the immobilized manganese catalyst onto the SiO2/Al2O3 was confirmed by
performing a series of consecutive experiments in which the used catalyst was filtered, washed
with fresh solvent, and employed without any further treatment in another run. The results shown
in Fig. 7 clearly prove that a slight decline of activity and selectivity occurs after fifth run.
Consequently, the results clearly suggest that the manganese catalyst efficiently catalyze
conversion of cyclohexene (84 %) with ca. 99 % selectivity to 2-cyclohexene-1-one under NHPI
and oxygen in benzonitrile as solvent. In order to confirm the heterogeneous nature of the
19
Page 21 of 54

manganese catalyst, a series of experiments were conducted in which the catalyst was filtrated
after 2, 4 and 6 h and the reactions were continued with the filtrate. Under these conditions the
reaction did not proceed at all, the data indicated that there was no leaching of the manganese
ions from the catalyst, therefore establishing the really heterogeneous nature of the catalyst.
3. 3. Oxidation of oximes
In this protocol, a novel useful procedure for the facile deoximation system by reusable Mn
supported catalyst and molecular oxygen as oxidant were studied (see Scheme 5). The
experiments were designed with acetophenone oxime as a model substrate. For optimizing the
conditions in oxidation of oximes, several factors have been investigated such as the amount of
catalyst, the effect of temperature, the amount of benzaldehyde and the effect of solvent. In this
reaction, the main product for oximes with electron-donating groups is carbonyl compounds. In
order to examine the effect of catalyst on deoximation reaction; the reaction was carried out in
the presence of various amounts of catalyst. The observations indicated that in the case of
absence of catalyst, the reaction is carried out slowly while in the presence of the optimum
amounts of the catalyst, the reaction is performed considerably. The maximum selectivity of
carbonyl compound was obtained with 0.15 g of catalyst. The conversion of the acetophenone
from 7% in the absence of the catalyst, to higher than 99% in the presence of the catalyst
indicates that the catalyst has an efficient and effective role in this reaction.
20
Page 22 of 54

For examining the role and effect of the solvent in the reaction, various solvents which are
suitable for radical reactions have been used [54]. Therefore, acetonitrile, benzene, toluene, and
cyclohexane solvents and water have been used. The results showed that in various solvents the
selectivity was 100 % and the maximum conversion decreased in the following order: toluene
(99.0) > acetonitrile (90.0) > benzene (83.0) > cyclohexane (15.0) > and water (0.0). The
reaction at the presence of coordinating solvents such as water made no reaction. It seems that
the donor electrons of this solvent had more ability to occupy the vacant space around the existed
metal in the catalyst, so this prevents coordinating from oxygen molecules [52]. In this work,
toluene provided the highest conversion and selectivity to acetophenone. Therefore, toluene as a
solvent plays an effective role in oxidation of oximes, because toluene was reluctant to undergo
free radical addition [53,54].
In oxidation of oximes, the presence of several amounts of benzaldehyde can also have a
determining role. Observations indicated that in the absence of benzaldehyde, the reaction has a
little progress even in a long time period, but in the presence of benzaldehyde a great conversion
was observed. This observation indicated that benzaldehyde was used as the oxygen transfer
agent in this oxidation reaction. In addition, when isobutyraldehyde was used as the only oxygen
transfer agent, great conversion was not observed (see Figure 8).
To estimate the reusability of the manganese supported catalyst, after the first performance
the catalyst was filtrated and washed with ether and dried at 120 ºC under vacuum and then used
21
Page 23 of 54

for the next performances under the same conditions. No apparent loss of activity and selectivity
was observed after 6th runs, confirming that the manganese catalyst has high stability during the
process. To further proofs that the reaction was catalyzed by the heterogeneous Mn-catalyst, we
added extra acetophenone oxime to the filtrate after the removal of the catalyst and found that no
more products were produced under the same conditions. In addition, the Mn ions absorbed on
the surface of the support did not contribute in the deoximation reaction. This observation
indicated that this reaction occurs due to the catalytic nature of the chemically immobilized of
Mn catalysts onto the Schiff base ligands. To further explore the utility of this catalytic system,
oxidation of several oximes (aldoximes and ketoximes) were also studied (see Table 4).
Aldehydes were obtained with excellent yields and the overoxidation to carboxcylicacid and
Beckmann rearrangement were not observed, which was another advantage of the proposed
system. Furthermore, steric structure almost has no effect on conversion and selectivity of oxime
(entry 14). The hydroxyl group in the ortho position of oximes (entry 12 and 13) exerts only an
partly inductive effect, but hydrogen bonding between the OH and the nitrogen atom in imine
will decrease the electron density of the carbon atom in the C=N. This factor probably influences
the reactivity of oxime. Consequently, the yield of this oxime against with other oximes is
slightly decreased, and the reaction time is longer[55].
In order to verify the kind of mechanism,2,6-di-tert-butyl-4-methylphenol(BHT), as a
scavenger of chemical radical was used. The reaction appeared to be radical processes since it
22
Page 24 of 54

was totally inhibited in the presences of BHT (5 mol %). It seems that oximes with different
groups (electron-donating and electron-drawing) go in various paths in the oxidation reaction
with oxygen over the catalyst .To further elucidate the reaction mechanism, series of experiments
were conducted, when the oxidation of acetophenoneoxime was conducted in the presence of
dioxygen and benzaldehyde, in the absence of manganese catalyst, the conversion of
acetophenone could only reach 9%. However, the conversion was remarkably increased by
adding 70 mg of manganese catalyst to the mixture (Table 5, 99%, entry 11). The results indicate
that manganese catalyst is crucial for the deoximation reaction. According to the proposed
mechanism, at the first Mn(III) ions and benzaldehyde generate an acyl radical (ph-CO·) , so the
radical formed is quickly trapped by dioxygen to give an acylperoxy radical (ph-COO·). The
acylperoxy radical acts as a carrier in a chain mechanism by reaction with another benzaldehyde
molecule to give the peroxybenzoic acid, and generating another acyl radical as well (equation 3)
[56-61]. The peroxybenzoic acid is assumed to play important roles, where peroxybenzoic reacts
with another Mn(III) catalyst molecule to generate high-valent cobalt intermediate (equation 4).
The color of the reaction mixture changed to tinge green which also indicated valence change of
Mn(III) to Mn(IV) [58-62]. Then, the intermediate attacks to the imine group of oxime through a
nucleophilic attack [61], thereby, the reaction of an oxo-catalyst radical and oximes, which is
proposed to generate a nitroso oxy radical species eventually loses NO to generate an aldehyde
and ketone. Via this process, oxime will be converted to its corresponding carbonyl compound.
23
Page 25 of 54

Although, the oxidation efficiency was not affected by electrondenoting group of substrate, the
presence of electron-drawing groups in oxime (like nitro group;entry 15 and 16) will change the
path of reaction (equation 6) so as the speed of protonation of hydroxyl group in oxime will be
more than the speed of oxo-catalyst attack to imine bond, consequently, it produces nitrile by the
elimination of water, the same result was observed in the literature [63-69] (see Scheme 6).
Mn and Co complexes are extensively used in oxidation of organic compounds as both a
catalyst and co-catalyst because of their oxidation–reduction property and ability to demonstrate
several valencies (Co: (II), (III); Mn: (II), (III), (IV) and (VII)) with easy electron transfer
between these states [70-72]. Recently, we reported the direct aerobic oxidation of ethylbenzene,
cyclohexene and various oximes to acetophenone, 2-cyclohexene-1-one and corresponding
carbonyl compounds using SiO2/Al2O3-supported cobalt catalyst in the presence of NHPI [73].
Manganese and cobalt spices in heterogeneous catalysts are typically main role in these
oxidation processes, wherein their catalytic effect lowers the activation energy of the
decomposition reaction of the alkylhydroperoxide. The synergistic effect of Co and Mn spices in
the aerobic oxidation of alkylbenzenes has been well documented [74]. The behaviours of the
Co(II) and Mn(III) ions may be a result of their different redox potentials and, in association with
this, their different behaviours in catalyzing hydroperoxide decomposition [75,76].Comparison
between performance of manganese and cobalt catalysts are shown in Table 5.
24
Page 26 of 54

4. Conclusions
A catalyst was developed based on the immobilization of manganese on SiO2/Al2O3 to
efficiently promote the aerobic oxidation of ethylbenzene, cyclohexene and oximes to
acetophenone, 2-cyclohexene-1-one and carbonyl compounds, under oxygen atmosphere.
Oxidation of alkyl aromatic and allylic site was resulted with the oxidant of NHPI and O2
without the use of any reductant. The catalytic performance was remarkably high in the
oxidation reaction. The effect of the catalyst amount, reaction time, reaction temperature and
effect of various solvents on the oxidation of ethylbenzene and cyclohexene were investigated.
The results indicated that Mn/NHPI facilitated the generation of the active species phtalimido-N-
oxyl (PINO) radical which further catalyzed the oxidation of ethylbenzene and cyclohexene.
Furthermore, we have reported a new and efficient methodology for the regeneration of
aldehydes and ketones from oximes. The extension of the method to other catalysts is currently
under investigation.
References
[1] L. Chen, B.D. Li, Q. X. Xu, D. B. Liu, Chin. Chem. Lett. In Press (2013).
[2] K. Sato, M. Aoki, J. Tagaki, R. Nayori, J. Am. Chem. Soc. 119 (1997) 12386-12387.
[3] R.S. Varma, R.K. Saini, Tetrahedron Lett. 39 (1998) 1481-1482.
[4] K.C. Gupta, A. Kumar Sutar, C. Lin, Coord. Chem. Rev. 253 (2009) 1926–1946.
[5] B. Tamami, S. Ghasemi, Appl. Catal. A 393 (2011) 242–250
.
25
Page 27 of 54

[6] D. Habibi, A. R. Faraji, Compets. Rendus. Chimie. 16 (2013) 888–896.
[7] R. Chakrabarty, D. Kalita, K.D. Brinichi, Polyhedron 26 (2007) 1239 ;
[8] W. Adam, F.G. Gelalcha, C.R. Saha-Mo¨ ller, V.R. Stegmann, J. Org. Chem. 65 (2000) 1915.
[9] F.C. Recupero, F.Punta, Chem. Rev. 107 (2007) 3800-3842.
[10] Y. Yoshino, Y. Hayashi, T. Iwahama, S. Sakaguchi, Y. Ishii, J. Org. Chem. 62 (1997)
9810-6813.
[11] Y. Ishii, K. Nakayama, M. Takeno, S. Sakaguchi, T. Iwahama, Y. Nishiyama, J. Org. Chem.
60 (1995) 3934-3935.
[12] Y.Ishii, S.Sakaguchi, T.Iwahama, Adv. Synth. Catal. 343 (2001) 393–427.
[13] Y. Ishii, J. Mol. Catal. A: Chem. 117 (1997), 123-137.
[14] P.J. Figiel, J.M. Sobczak, J.J. Ziolkowski, Chem. Commun. (2004) 244-245.
[15] P.J. Figiel, J.M. Sobczak, New. J. Chem. 31 (2004) 1668-1673.
[16] T. Iwahama, K. Soyojyo, S. Sakaguchi, Y. Ishii, Org. Proc. Res. Dev. 2 (1998) 255-260.
[17] B.B. Wentzel, M.P.J. Donners, P.L. Alsters, M.C. Feiters, R.J.M. Nolte, Tetrahedron. 56
(2000) 7797-7803.
[18] F. Yang, J. Sun, R. Zheng, W. Qiu, J. Tang, M. He, Tetrahedron. 60 (2004) 1225-1228.
[19] B. Ferg, L. Hau, H. Yang, Y. Hu, H. Li, X. Zhao, Catal. Commun. 10 (2009) 1542-1546.
[20] S. Evans, J.R.L. Smith, J. Chem. Soc, Perkin Trans. 2 (2001) 174-180.
[21] R. Wang, B. Gao, W. Jiao, Appl. Surf. Sci. 255 (2009) 4109-4113.
[22] S. Vetrivel, A. Pandurangan, Appl. Catal. A: Gen. 264 (2004) 243-252.
[23] Y. Ishii, T. Iwahama,S. Sakaguchi, K. Nakayama, Y. Nishiyama, J. Org. Chem. 61 (1996)
4520-4526.
[24] R.A. Sheldon, I.W.C.E Arends, Adv. Synth. Catal. 346 (9-10) (2004) 1051-1071.
26
Page 28 of 54

[25] R.A. Sheldon, I.W.C.E Arends, J. Mol. Catal. A: Chem. 251 (1-2) (2006) 200-214.
[26] F. Minisci, F. Recupero, A. Cecchetto, C. Gambrotti, C. Punto, R. Faletti, R. Paganelli, G.F.
Pedulli, Eur. J. Org. Chem. 1 (2004) 109-119.
[27] R. Amoroti, M. Lucarini, V. Mugnaini, G.F. Peduli, F. Minisci, F. Recupero, F. Fontana, P.
Astolfi, L. Greci, J. Org. Chem. 68 (5) (2003) 1747-1754.
[28] F. Menisci, F. Recupero, G.F. Pedulli, M. Lucarini, J. Mol. Catal. A: Chem. 204-205 (2003)
63-90.
[29] I. Hermans, P. Jacobs, J. Peeters, Phys. Chem. Phys. 9 (6) (2007) 686-690.
[30] G. Dasiva, J.W. Bozzelli, J. Phys. Chem. 111 (2007) 5760-5765.
[31] I. Hermans, P.A. Jacobs, J. Peeters, Phys. Chem. Chem. Phys. 10 (2008) 1125-1132.
[32] I. Hermans, L. Vereecken, P.A. Jacobs, J. Peeters, Chem. Commun. 9 (2004) 1140-1141.
[33] N. Koshino. Y. Cai, J.H. Epenson, J. Phys. Chem. A. 107 (2003) 4262-4267.
[34] M. Arshadi, M. Ghaiaci, Appl. Catal, A: Gen. 399 (2011) 75-86.
[35] Y. Ishii, S. Sakaguchi, Catal.Today. 117 (2006) 105–113.
[36]B. B. Wentzel, M. P. J. Donners, P. L. Alsters, M. C. Feitersa, R.J. M. Nolte, Tetrahedron.
56 (2000) 7797-7803.
[37]F. Rajabi, R. Luque, J. H. Clark, B. Karimi, D. J. Macquarrie, Catal. Commun. 12 (2011)
510–513.
[38]Y. Yoshino, Y. Hayashi, T. Iwahama, S. Sakaguchi, Y. Ishii, J. Org. Chem. 62 (1997) 6810-
6813.
[39] D. Habibi, A.R. Faraji, M. Arshadi, J.L.G. Fierro, J.Mol.Catal A: Chem. 372 (2013) 90– 99.
[40] D. Habibi, A.R. Faraji, Appl. Surf. Science. 276 (2013) 487– 496.
[41] Z. Ozaydin, S.Yasyerli, G. Dogu, Ind. Eng. Chem. Res. 47 (2008) 1035-1042.
27
Page 29 of 54