Dicationic ionic liquids/heteropoly acid composites as heterogeneous catalysts for cyclohexene oxidation with molecular oxygen under solvent-free condition: Insights from theory and experiments

Article abstract of DOI:10.1016/j.apcata.2017.06.025

This study describes solvent-free oxidation of cyclohexene using dicationic ionic liquids (DIL)/heteropoly acid (HPA) composites with molecular oxygen. Structural characterization of the DIL/HPA composites was carried out by different techniques such as ICP, FT-IR, ¹H NMR, BET, TGA, XRD, FESEM, EDX and TEM. Characterization results confirmed that heteropoly acids i.e., tungstophosphoric (H₃PW₁₂O₄₀, TPA) and molybdophosphoric acid (H₃PMo₁₂O₄₀, MPA) were incorporated with DILs without distortion of the Keggin structure of the heteropoly acids. Two different dicationic ionic liquids with linear and aromatic linkage, DIL-L and DIL-A, were prepared and applied for synthesis of the composites. Effective parameters including the effect of temperature, time, oxygen pressure, amount of the selected catalyst, i.e., DIL-L/tungstophosphoric acid was thoroughly investigated. The catalyst was found to be stable and recyclable. The main product of the reaction was cyclohexene-2-one (Cy-one) at 70?°C and 15?bar oxygen pressure after 6?h. Additionally, theoretical calculations at the PM6 semi-empirical method was performed on DILs/HPA complexes with different number of DIL molecules to explore the capacity of HPA in interaction with DILs and obtained the orientations of DILs around the Keggin oxoanion of the HPA. The calculations showed that the maximum number of DIL molecules around the [H₂PW₁₂O₄₀]^? oxoanion was five and the interaction of DIL-L withthe oxoanion was stronger than that of DIL-A which was in agreement with the experimental data.

Full text of DOI:10.1016/j.apcata.2017.06.025

Accepted Manuscript
Title: Dicationic Ionic Liquids/Heteropoly Acid Composites
as Heterogeneous Catalysts for Cyclohexene Oxidation with
Molecular Oxygen under Solvent-free Condition: Insights
from Theory and Experiments
Authors: Fahime Nezampour, Mehran Ghiaci, Hossein
Farrokhpour
PII:
S0926-860X(17)30274-0
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2017.06.025
APCATA 16287
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
22-3-2017
31-5-2017
19-6-2017
Please cite this article as: Fahime Nezampour, Mehran Ghiaci, Hossein
Farrokhpour, Dicationic Ionic Liquids/Heteropoly Acid Composites as Heterogeneous
Catalysts for Cyclohexene Oxidation with Molecular Oxygen under Solvent-
free Condition: Insights from Theory and Experiments, Applied Catalysis A,
Generalhttp://dx.doi.org/10.1016/j.apcata.2017.06.025
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.

Dicationic Ionic Liquids/Heteropoly AcidComposites as Heterogeneous
Catalysts for Cyclohexene Oxidation with Molecular Oxygen under Solvent-
free Condition: Insights from Theory and Experiments
Fahime Nezampoura, Mehran Ghiacia,*and Hossein Farrokhpoura
aDepartment of Chemistry, Isfahan University of Technology, Isfahan,8415683111,Iran
*mghiaci@cc.iut.ac.ir
Graphical Abstract
Highlights
 Dicationic ionic liquids/heteropoly acid composites as heterogeneous catalysts.
 Molecular oxygen and cyclohexene oxidation under solvent-free conditions.
 DIL-L/W composite as the best catalyst and 96% conversion of cyclohexene.
 Semi-empirical method and the interaction energy of TPA with DILs.
1

Abstract
This study describes solvent-free oxidation of cyclohexene using dicationic ionic liquids
(DIL)/heteropoly acid (HPA) composites with molecular oxygen. Structural characterization of
1
the DIL/HPA composites was carried out by different techniques such as ICP, FT-IR, H NMR,
BET, TGA, XRD, FESEM, EDX and TEM. Characterization results confirmed that heteropoly
acids i.e., tungstophosphoric (H3PW12O40, TPA) and molybdophosphoric acid (H3PMo12O40,
MPA) were incorporated with DILs without distortion of the Keggin structure of the heteropoly
acids. Two different dicationic ionic liquids with linear and aromatic linkage, DIL-L and DIL-A,
were prepared and applied for synthesis of the composites. Effective parameters including the
effect of temperature, time, oxygen pressure, amount of the selected catalyst, i.e., DIL-
L/tungstophosphoric acid was thoroughly investigated. The catalyst was found to be stable and
recyclable. The main product of the reaction was cyclohexene-2-one (Cy-one) at 70 °C and 15
bar oxygen pressure after 6 h. Additionally, theoretical calculations at the PM6 semi-empirical
method was performed on DILs/HPA complexes with different number of DIL molecules to
explore the capacity of HPA in interaction with DILs and obtained the orientations of DILs
around the Keggin oxoanion of the HPA. The calculations showed that the maximum number of
DIL molecules around the [H2PW12O40]- oxoanion was five and the interaction of DIL-L withthe
oxoanion was stronger than that of DIL-A which was in agreement with the experimental data.
Keywords:
Oxidation;
Cyclohexene;
Heteropolyacid;
Dicationic
ionic
liquid;
Bis(imidazolium)compounds; Heterogeneous catalyst.
2

1. Introduction
Oxidation of unsaturated hydrocarbons on the allylic and the double bond positions are
very important transformations in synthetic organic chemistry because of the importance of the
products as valuable commercial intermediates [1-3].
Cyclohexene as an important petrochemical resource, in the oxidation process could
undergo provoking olefinic oxidation at the double bond in addition to allylic oxidation.
Therefore, it is reasonable to expect that due to the tendency of the oxidant for the allylic
position as well as for the double bond position in the reaction medium, the reaction will end up
with the formation of the allylic oxidation products, i.e., 2-cyclohexen-1-one (Cy-one) and 2-
cyclohexen-1-ol (Cy-ol) [1,4-8], or epoxidation products i.e., cyclohexene epoxide and 1,2-
cyclohexanediol (CyH-diol) [9,10]. It is known that the type of oxidant used in the reaction could
conduct the reaction to the specific products [11-13]. It is believed that by using hydrogen
peroxide, NaOCl, KHSO5 and PhIO the oxidation occurs at the double bond position, and by
using the bulky oxidants such as tert-butyl hydroperoxide (TBHP) because of the steric
hindrance of the bulky tert-butyl group, the oxidation would occur at the allylic position of the
cyclohexene [14-16]. Mechanistically, there is a hypothesis that allylic oxidation occurs through
a free-radical chain reaction route [17]. Reported data in the literatures [18-20] have emphasized
that the metallic species which are in a low oxidation state follow up the free-radical pathway.
On the other hand, the oxometallic reagents with higher oxidation states, such as the catalyst
reported in this work, i.e., W=O or Mo=O are prone to facilitate the double bond oxidation
because these type of transition metals by having empty d-orbitals could accept electron pairs
from oxidant (hydrogen peroxide, molecular oxygen, etc) and increasing the electrophilic nature
of the oxidant and becoming liable to be attacked by the π-electrons of the olefinic substrate [21].
3

It should be mentioned that in the cases where one observes the allylic and double bond
oxidation products both mechanisms could occur simultaneously.
The basic feature of the parameters that govern the regioselectivity of a catalyst is a
fundamental question. To best of our knowledge, in the oxidation of cyclohexene the key
reactions to be mentioned are the allylic and double bond oxidation. Different catalysts give
different ratio of allylic and double bond oxidation products, and the ratio of C-H/double bond
oxidation is more than one in most cases. The present paper addresses the regioselectivity
problem by shifting this ratio to one extreme, i.e., predominantly oxidizing the allylic position. In
this respect, we decided to monitor the effect of environment of Keggin oxoanion on its catalytic
performance [18]. It should be mentioned that many Keggin oxoanion-based ionic hybrid
composites have been used as heterogeneous catalysts. For example, Wang and his co-workers
used a family of HPA/dicationic-IL composites as reusable catalysts for hydroxylation of
benzene [22] and esterification [23]. Moreover, Leng, et al., have used a mesoporous amphiphilic
HPA/dicationic-IL as a highly efficient epoxidation catalyst [24]. However, dissolution of HPA-
IL composites in some common oxidants such as H2O2 is a drawback for these hybrid
composites as desirable liquid-solid heterogeneous catalysts in oxidation reactions [25, 26].
In this context, we had this in mind to disperse polyoxometalates of the used heteropoly
acids without losing a large extent of their oxidative sites by supporting them on a solid support
[19,20,27]. Therefore, we precipitated the water soluble tungstophosphoric and
molybdophosphoric acids with two different dicationic ionic liquids (DILs), 1 and 2. We think
that a uniform dispersion of the oxoanions between the DILs and increasing their hydrophobicity
might be considered as the main reason responsible for the high catalytic performance of the
prepared catalysts.
4

Our semi-empirical calculations have shown that each oxoanion unit could hold
maximum of five DIL molecules electrostatically around itself and the interaction of DIL-L with
the [H2PW12O40]-oxoanion was stronger than that of DIL-A.
2.Experimental
1. Materials
Triphenylphosphine was obtained from Fluka. Bromine and 1,4-dibromobutane were
purchased from Sigma-Aldrich. All other necessary chemicals such as 1,4-
bis(hydroxymethyl)benzene, silica-gel, 1-methylimidazole, tungstophosphoric acid (TPA),
molybdophosphoric acid(MPA) and cyclohexene were purchased from Merck Company.
Molecular oxygen (99.99 %) was obtained from Argon Company (Iran). All solvents used in this
study such as CCl4, EtOH and acetonitrile were analytical grade.
2.2. Characterization Techniques
The products were characterized by FT-IR, 1H NMR, GC, GC-MS, FESEM-EDX, TEM,
BET, ICP-OES and XRD techniques which they are described in the supporting information.
2.3. Preparation of the Catalysts
2.3.1. Synthesis of 1,4-Bis(bromomethyl)benzene, 3,3'-(Butane-1,4-diyl)bis(1-methyl-1H-
imidazol-3-ium) bromide and 3,3'-(1,4-Phenylenebis(methylene))bis(1-methyl-1H-imidazol-3-
ium) bromide
The procedures for synthesis of 1,4-Bis(bromomethyl)benzene,3,3'-(Butane-1,4-
diyl)bis(1-methyl-1H-imidazol-3-ium) bromide and 3,3'-(1,4-Phenylenebis(methylene))bis(1-
methyl-1H-imidazol-3-ium) bromide have been described in the supporting information, and
1
structures of both products were confirmed by FT-IR and H NMR techniques (Figs S1-S3).
5

Henceforth, the 3,3'-(butane-1,4-diyl)bis(1-methyl-1H-imidazol-3-ium) dibromide and 3,3'-(1,4-
phenylenebis(methylene))bis(1-methyl-1H-imidazol-3-ium) dibromide are denoted as DIL-L (1)
and DIL-A (2), respectively (Scheme 1).
Scheme 1. Preparation of the (a) DIL-L (1) and (b) DIL-A (2).
2.3.2. Preparation of 3,3'-(Butane-1,4-diyl)bis(1-methyl-1H-imidazol-3-ium)/Heteropoly acid
and
3,3'-(1,4-Phenylenebis(methylene))bis(1-methyl-1H-imidazol-3-ium)/Heteropoly
acid
Composites [DIL-L/W(Mo) and DIL-A/W(Mo)]
For preparation of DIL-L/tungstophosphoric acid (DIL-L/TPA), and DIL-
L/molybdophosphoric acid (DIL-L/MPA) catalysts, 0.1 g (0.263 mmol) of DIL-L was dissolve
in 2 mL of DI water and an aqueous solution of 0.076 g (0.026 mmol) [or 0.15 g (0.05 mmol), or
0.3 g (0.1 mmol)] TPA or 0.048 g (0.02 mmol) [or 0.096 g (0.05 mmol)] MPA in 2 mL of DI
water was slowly added to it. Then, the solid was separated by filtration, dried at room
6

temperature and renamed hereafter as DIL-L/W(x) or DIL-L/Mo(x), respectively; the number in
parenthesis shows the weight percentage of the metal in the catalyst.
To prepare DIL-A/tungstophosphoric acid (DIL-A/TPA) and DIL-A/molybdophosphoric
acid (DIL-A/MPA) catalysts, 0.1 g (0.234 mmol) of DIL-A was dissolved in 2 mL of DI water
and an aqueous solution of 0.067 g (0.026 mmol) [or 0.134g (0.05 mmol)] TPA or 0.085 g (0.05
mmol) MPA in 2 mL of DI water was slowly added to it. The resulting solid was separated by
filtration, dried at ambient temperature and renamed hereafter as DIL-A/W(x) or DIL-A/Mo(x),
respectively; the number in parenthesis shows the weight percentage of the metal in the catalyst
(Scheme 2).
2.3.3. General Procedure for Oxidation of Cyclohexene
A 50 mL titanium batch reactor was charged with cyclohexene (2 mL) and the catalyst
(20 mg). The desired pressure (15 bar) was provided with oxygen and the selected temperature
was set [28]. Reaction was carried out under solvent-free condition at the selected temperature.
After performing the reaction for the preset time and cooling the reactor, the mass balance was
checked. Then, after separating the catalyst products were analyzed quantitatively by GC using
p-xylene as internal standard. The yield of 2-cyclohexene-1-one was defined as the ratio of
theweight of 2-cyclohexene-1-one, as determined by GC, to the weight of 2-cyclohexene-1-one
theoretically produced upon oxidation of cyclohexene; the products were identified by GC-Mass
(Scheme 3).
7

Scheme 2. Synthesis of DIL-L/W(Mo) and DIL-A/W(Mo).
Scheme 3. Oxidation of cyclohexene in the presence of DIL-L/W composite.
3. Results and Discussion
8

3.1. Characterization of the Synthesized Catalysts
In the next sections we will show that the DIL-L/W(33) was the preferred catalyst on the
basis of the conversion, selectivity and cost-effective. Therefore, we mainly concentrated on
characterization of this catalyst. The DIL-L/W(33) catalyst was prepared by electrostatic
interactions between the DIL-L and the oxoanion of the TPA [PW12O40]n-. Scheme 2 shows the
synthetic route for formation of the [DIL]n2+[HmPM12O40]n-. The metal content of the catalyst
before and after the catalytic reaction was measured by atomic absorption spectroscopy ICP-
OES; as shown in the reusability experiments and in agreement with data of FESEM and XRD
analysis, the catalyst probably reaches to its best TPA/DIL ratio, and in this process loses a
fraction of ionic liquid which affects on the reusability of the catalystby becoming less
hydrophilic and in a way its contact with cyclohexene decreases. The EDX data and EDX
mapping show the presence of metal ions, phosphorous and nitrogen in the fresh and reused
catalyst (Fig. 1).
The FESEM images of the fresh DIL-L/W(33) catalyst and after 4 runs are shown in Fig.
1 with four different magnifications and particle size distribution histogram of 200 nm scale.
According to these images, the fresh catalyst (Fig. 1a) is composed of spherical nanoparticles
with an average size of around 35 nm. The morphology of the catalyst after 4 runs completely
changes, and in agreement with XRD of the reused catalyst (Fig. 3g), crystallinity of the catalyst
improved to a large extent.
9

Fig. 1. FESEM images of (a) fresh DIL-L/W(33) and reused DIL-L/W(33) at magnification of
200 nm, 500 nm and 1 μm and EDX confirmation of the presence of W element and elemental
mapping images (inset).
10

The companionship of phosphotungstate anion with DIL-L and formation of the DIL-L/W as
catalyst was confirmed by EDX analysis. According to the elemental mapping analysis the
elements of C, N, O, Br, P and W have been discovered with a homogeneous distribution of P
and W on the catalyst (Fig. 1).
The significant stretching vibrations of the prepared catalyst and its components have
been helpful for identification of the catalyst. Fig. 2 shows the IR spectra of the DIL-L, TPA,
DIL-L/W(33) and reused DIL-L/W(33) catalyst after four runs; we have also measured the IR
spectra of the DIL-A/W(Mo) composites which are shown in Fig. S4. In the spectrum of DIL-L
the broad peak at 3500 cm-1 is attributed to the –OH stretching of the moisture presents in the
hydrophilic DIL-L, and the peaks at 3145 and 3073 cm-1 are assigned to the C-H stretching of the
imidazolium rings. Also the peaks in the region of 2862-3000 cm-1 are attributed to the aliphatic
stretching vibrations. Moreover, the peaks at 1632, 1577 cm-1 and 1158 cm-1 were ascribed to the
breathing vibrations of C=N and C-H out-of-plane bending vibrations of the imidazolium rings,
respectively (Fig. 2a). TPA has four characteristic bands which are appearedinthe range of 800-
1100 cm-1. These peaks have been observed at 804, 891, 982 and 1079 cm-1that are attributed to
edge shared oxygen (W−Oc−W), corner shared asymmetric oxygen(W−Ob−W), terminal
asymmetric oxygen (W=Od) and terminal asymmetric oxygen (P−Od), respectively (Fig. 2b)
[24]. For DIL-L/W(33) catalyst these peaks are located at 803, 895, 976 and 1078 cm-1, so that
the red shift observed in the peak corresponding to W=Od bond probably suggests the existence
of an electrostaticinteraction between oxygen of the W=Od bond in the Keggin oxoanion with the
positive charge of the imidazolium ring of the surrounding ionic liquids [29, 30].
11

Fig. 2. FT-IR spectra of (a) DIL-L, (b) TPA, (c) DIL-L/W, (d) reused DIL-L/W, (e) MPA and (f)
DIL-L/Mo.
The XRD patterns of the pure H3PW12O40 (TPA), DIL-L and DIL-L/W catalysts with
different molar ratios of TPA are illustrated in Fig. 3. Clearly one could see that H3PW12O40
displayed a set of well resolved sharp diffraction peaks, Fig. 3b. As mentioned, we prepared
three different samples of the catalyst with different ratios of DIL-L/TPA (33, 46 and 57 wt% of
the metal), Figs.(3d-f). As can be seen, obviously by decreasing the amount of TPA in the
composite the crystallinity of the composite increases, and more interesting in Fig. 3g which
belongs to the reused catalyst the crystallinity has improvedto a large extent; .What do we
understand from these changes? We think by decreasing the amount of TPA in the composite,
12

the ratio of oxoanion and the ionic liquid reaches to the optimum value to form a
thermodynamically stable crystal.
Fig. 3. XRD pattern of (a) DIL-L (support), (b) TPA, (c) a+b, (d) DIL-L/W(33), (e) DIL-
L/W(46), (f) DIL-L/W(57) and (g) reused DIL-L/W(33).
The TEM images of the DIL-L/W(33) catalyst sample are presented in Fig. 4 which
shows a 2D microstructure composed of nanoparticles. The approximately spherical
nanoparticles were compact and did not show any porous network.
13

Fig. 4. TEM images of DIL-L/W(33) at different magnifications.
The specific surface area of 30 (m2/g), pore volume of 0.126 (cm3/g) and pore diameter
of 2.5 nm for the DIL-L/W(33) catalyst was obtained using nitrogen adsorption-desorption
isotherm (Table 1) which indicates the presence of macropores in the prepared catalyst, and
reasonable surface area of the catalyst might be due to nano-size of the crystalline composite
(Fig. 5). Moreover, the nanoparticles became smaller after a few runs of oxidation reaction
because of leaching of TPA from the catalyst.
Table 1. Textural Properties of DIL-L/W(33)
Catalyst
SBET (m2/g)
Vpore (cm3/g)
Dpore (nm)
2.5
DIL-L/W(33)
30
0.126
14

Fig. 5. N2 adsorption ( ) and desorption ( ) isotherm of DIL-L/W(33) and dBJH-Plot pore size
distribution (inset).
The thermogravimetric curve of the DIL-L/W(33) catalyst has been depicted in Fig. S5.
3.2. Theoretical Calculations
In this part, theoretical calculations were used to explore the capacity of the oxoanion to
be surrounded by the DIL molecules. The TPA is a strong acid with three acidic hydrogen atoms
which could produce three anionic forms such as TPA-, TPA2- and TPA3-in aqueous solution.
The pKa values of the TPA are 1.6, 3.54 and 7.16, respectively [31]. Therefore, it is expected that
the dominant anionic form of TPA in water is TPA-. There were a few questions that we were
encourage for performing the following calculations in this work including (1) to obtain the
correct orientation of DILs around TPA- oxoanion (2) calculating the maximum number of
interacting DILs and (3) predicting the change in the interaction energy per each DIL unit that
interacting with the TPA- oxoanion. The semi-empirical PM6 method [32] was used for the
15

optimization of the structures and the calculations of the interaction energies. The reason for
choosing of this method was the size of TPA- oxoanion and the size of the DIL molecules. The
composite of TPA- with different number of DIL molecules were optimized, separately. Fig. 6
displays the optimized structures of [(DIL-L)n/TPA-; n=4, 5 and 7] composites. The other
optimized structures of the [(DIL-L)n=1-3/TPA-] composites have been shown in Fig. S6. The
optimized structures of the [(DIL-L)n=1-3/TPA-] show that the DIL-L molecule prefers to have a
conformer with the two imidazolium rings near the TPA- oxoanion to create desire electrostatic
interactions with the oxygen atoms of the oxoanion. By adding the 4th DIL-L to [(DIL-L)3/TPA-]
to form [(DIL-L)4/TPA-] it also acts like other DILs. The optimized structure of [(DIL-L)5/TPA-]
shows that two of the DIL-Ls have vertical positions relative to the TPA-. The trend of increasing
the number of DILs was continued and our calculations showed that by increasing the number of
DILs molecules the oxoanion loses some of its electrostatic interactions with the previously
anchored DILs to open room for other DIL molecules while keeping three DIL molecules very
close to it. The optimized structure of the [(DIL-L)7/TPA-] shows four DIL-Ls in vertical
orientation relative to TPA- and the other three DIL molecules had maximum interactions with
oxoanion (Fig. 6c). It is interesting to notice that the distance between two of them and the TPA-
oxoanion is considerably higher than the others.
16

(a)
3.5Å
2.9 Å
(b)
(c)
4.9 Å
2.9 Å
11.3 Å
9.7 Å
Fig. 6. The optimized structures of [(DIL-L)n/TPA-; (a) n=4, (b) n=5 and (c) n=7].
17

Similar calculations were performed on [(DIL-A)n=1-5/TPA-] structures and their
optimized structures have been demonstrated in Fig. S7. It should be mentioned that the presence
of the aromatic linkage between two imidazolium rings in the DIL-A reduces the flexibility of
the molecule for the interaction with the TPA-. Comparison of the optimized structure of the
[(DIL-L)2/TPA-] with that of [(DIL-A)2/TPA-] confirmed this prediction. The orientation of the
DILs relative to TPA-oxoanion in the optimized structure i.e., [(DIL-A)4/TPA-] is considerably
different from that observed in the optimized structure of the [(DIL-L)4/TPA-] composite. One of
the DILs has vertical orientation relative to TPA- in [(DIL-A)4/TPA-] unlike to what is seen in
the [(DIL-L)4/TPA-]. Generally, electrostatic interaction of the DILs with TPA-in the vertical
position was lower than this interaction with the conformer that the two imidazolium rings facing
the TPA- oxoanion. Moreover, similar calculations were performed on TPA2- and TPA3- and the
same optimized structures were obtained for the interaction of different numbers of DIL-L or
DIL-A molecules which show that the optimized structures are independent of the charge on
TPA oxoanion (not reported).
In these calculations the interaction energy between the TPA- and different numbers of
DILs was obtained using the following equation:
E=Ecomplex –ETPA--EDILs
Where Ecomplex is the total electronic energy of TPA- interacting with DILs, ETPA- is the electronic
energy of the oxoanion and EDILs is the total electronic energy of DILs, interacting with each
other in the absence of TPA-. Fig. 7 demonstrates the variation of the interaction energy per each
DIL unit (E/n) versus the number of DILs for both DIL-L and DIL-A. It can be seen that by
increasing the number of DILs, the value of (E/n) becomes more positive for both DIL-L and
DIL-A molecules. In regard to Fig. 7, there are two important points that should be emphasized:
18

(1) the interaction of TPA- with DIL-L is stronger than that of DIL-A (2) the interaction of TPA-
reaches to a saturated value of around n=6 and n=5 for [(DIL-L)n/TPA-] and [(DIL-A)n/TPA-],
respectively.
. DIL-L/TPA-
. DIL-A/TPA-
Fig. 7. Variation of the interaction energy/DIL unit (∆E/n).
3.3. Catalytic Performance of the Prepared Catalysts
The effect of temperature on the oxidation of cyclohexene over the various prepared
catalysts was studied by changing the temperature from 60 ºC to 100 ºC while the other
parameters were fixed. The results of oxidation with molecular oxygen in regard to conversion
and selectivity are summarized in Fig. 8. At 60 ºC, the catalysts containing molybdenum did not
react with cyclohexene although the catalysts containing tungsten show a moderate conversion
19

and selectivity to 2-cyclohexene-1-one; less than 2% cyclohexene oxide was detected at this
temperature. In the other extreme, i.e., at 100 ºC the catalysts having phosphomolybdate
demonstrated an excellent conversion and selectivity to 2-cyclohexene-1-one but the catalysts
containing TPA undergoover-oxidation. This could be due to the higher activity of the
tungstophosphoric acid among the Keggin series of HPAs. Moreover, at 70 ºC only the DIL-L/W
shows considerable selectivity to the main product. At 80 ºC all of the catalysts exhibited a high
conversion and selectivity.
Fig. 8. Influence of the reaction temperature on the cyclohexene oxidation. Reaction conditions:
cyclohexene 2 mL; different catalysts DIL-L/W(Mo)(46) and DIL-A/W(Mo)(46) 20 mg; 15 bar
O2; 6 h under solvent free.
20

As shown in the Fig. 9, the time of the reaction evidently had a significant effect on the
DIL-L/W(Mo)(46) and DIL-A/W(Mo)(46) catalytic selectivity of the reaction although the
conversion was high in all of the reactions specially for DIL-L/W and DIL-A/W catalysts. When
employing the DIL-L/W(46) catalyst for 4 h, a desirable conversion of cyclohexene was
achieved with 98% total selectivity for 2-cyclohexene-1-one such that the yield of this product
was over 90%. By increasing the time of reaction to 8 h, clearly the DIL-L/W catalyst exhibited
the best performance. Finally, on the basis of the presented data the 6 h was chosen as the
optimum time for the reaction.
Fig. 9. Influence of the reaction time on the cyclohexene oxidation. Reaction conditions:
cyclohexene 2 mL; different catalyst (DIL-L/W(Mo)(46) and DIL-A/W(Mo)(46) 20 mg; 15 bar
O2; under solvent free.
21

On the basis of the presented data we did choose the DIL-L/W as the selected catalyst. In
this respect to demonstrate the effect of the weight percentage of the phosphotungstate in the
DIL-L/W catalyst for oxidation of cyclohexene, the amount of TPA was varied from 33 wt% to
57 wt% while the other parameters were kept constant. The results on the conversion of
cyclohexene and selectivity to 2-cyclohexene-1-one are summarized in Table 2. Clearly, by
increasing the percentage of TPA in the composite of the DIL-L/W catalyst the conversion starts
decreasing which might be related to agglomeration of the catalyst.
Fig. 10 demonstrates the effect of the initial oxygen pressure on the oxidation of
cyclohexene over the DIL-L/W(33) catalyst. It is clear that by increasing the initial O2 pressure
conversion increased but the selectivity to the main product decreased to some extent. In this
respect, one could imagine that the reaction site has changed toward the oxidation of the double
bond position. Also by increasing the pressure from 15 to 20 bars, conversion decreased which
might be ascribed to the “saturation” of the exposed catalytic sites at higher O2 pressures
[5,33,34]. Therefore, by considering the conversion of cyclohexene and the selectivity for 2-
cyclohexene-1-one, 15 bar was chosen as the optimal initial oxygen pressure.
22

Table 2. Influence of the weight percent of tungsten in DIL-L/W and reaction media on
cyclohexene oxidation
Selectivity (%)
T
W
Conv.
Entry Catalyst
(°C) (wt%) .(%)
Cy-ol
CyH-diol Cy-one
1
2
3
4
5
DIL-L
80
80
80
80
80
-
34
84
89
94
21
2
-
2
96
94
83
95
45
DIL-L/W
DIL-L/W
DIL-L/W
DIL-L/W*
57
46
33
33
6
-
17
-
5
14
41
33
6
DIL-L/W** 80
DIL-L/W 70
N.R.
*
*
*
33
7
96
-
14
86
Reaction condition: cyclohexene 2 mL; catalyst 20 mg; 15 bar O2; 6 h; 80 °C and 70 °C under
solvent free condition.
*In the presence of acetonitrile 5 mL
**In the presence of acetic acid 5 mL
Fig. 10. Influence of the O2 pressure on the cyclohexene oxidation. Reaction conditions:
cyclohexene 2 mL; DIL-L/W(33) 20 mg; 70 °C under solvent free; 6 h.
The effect of amount of catalyst for oxidation of cyclohexene was studied by varying the
amount of DIL-L/W(33) catalyst from 10-30 mg while the other parameters were fixed (Fig. 11).
23

Fig. 11. Influence of the catalyst amount on the cyclohexene oxidation. Reaction conditions:
cyclohexene 2 mL; 15 bar O2; 6 h under solvent free,70 °C.
The results reveal the increase of conversion when the amount of catalyst was 20 mg and
by enhancing its amount to 30 mg no remarkable change was observed in the conversion of the
reaction. Therefore, 20 mg of the catalyst was taken as optimum amount of the catalyst.
The effect of solvent on the oxidation of cyclohexene over the DIL-L/W catalyst was
studied to reach to a better understanding of the catalytic system. The used HPAs are soluble in
water but they precipitate in the presence of the dicationic ionic liquids prepared in this work
which it could mean that in this trend the hydrophilicity of the HPA has decreased to a large
extent and perhaps should be more compatible with the cyclohexene as a hydrophobic substrate.
If this is true one would expect that by increasing the polarity of the media the interaction
between the substrate and the catalyst should decrease appreciably. In this respect, acetonitrile
and glacial acetic acid were chosen as the media of the oxidation reaction. In each case, the
oxidation reaction was performed by 5 mL solvent and 2 mL cyclohexene while other parameters
were fixed and similar to solvent-free reactions. As shown in Table 2 (entries 5 and 6) the
reaction did not occur in acetic acid, and the conversion and selectivity were unsatisfactory when
the reaction was performed in acetonitrile.
24

And finally, in this catalytic reaction the formation of peroxo species of the metal (W or
Mo) is proposed as the key intermediate in the first step. In the second step, coordination of the
double bond to metal causes oxidation of the cyclohexene in different positions. Perhaps the
water that might be present in the catalyst converts the produced cyclohexene epoxide with the
help of acidic property of the TPA to 1,2-cyclohexanediol.
3.4. Reusability of the Catalyst
The reusability of the catalyst was examined by recovering the catalyst from the reaction
mixture and used in the next run. Before using the catalyst in the next run, it was washed with
ethanol, dried at60 °C for overnight and then used as the catalyst under similar conditions. The
DIL-L/W(33) catalyst was recycled for 4 runs and as shown in Fig. 12, the conversion of the
reaction in the4th run decreased to 70%. As mentioned in section 3.1, the catalyst after 4 runs
loses some of the ionic liquid component and reaches to a better crystallinity and in this respect
the activity of the catalyst decreased appreciably which could not be due to the leaching of
tungsten but might be attributed to structural changes of the catalyst.
100
75
50
25
0
1
2
3
4
Runs
Fig. 12. Reusability of the DIL-L/W(33) catalyst.
25

Different catalytic systems such as Au/Silica [5], Au/HNTs [35], Cr-MCM-41 [36], PS-
DA-Cu [37] and Fe3O4@SiO2-CoO [38] are few examples of the catalysts which have been
reported for oxidation of cyclohexene under solvent-free conditions with O2 as the oxidant. The
data in the Table 3 compare the activity of the DIL-L/W catalyst and other catalysts reported in
the literature. Perhaps, the catalysts exhibited in the present work could be appropriate catalysts
for producing 2-cyclohexene-1-one.
Table 3. Comparison of DIL-L/W and DIL-A/W catalytic performance with some previously
reported catalyst in cyclohexene oxidation.
PO2
Temp. Reaction time Conversion
Entry
Catalyst
Ref.
(bar) (°C)
(h)
15
(%)
1
Cu-MOF a
1
80
80
80
80
80
80
75
70
70
70
70
14.6
[39]
[6]
2
GNPs/TChDb
17
8
8
87
3
Au-bentonite c
Au-silica c
10
8
53
[5]
4
13
4
92
[5]
5
Au/La-OMS-2 d
Au/HNTs e
24
12
6
48
[40]
[35]
[38]
[41]
[42]
[36]
[37]
6
4
29.5
96.5
91.4
68.6
52
7
Fe3O4@SiO2-CoO f*
P-Salen-BsdB18C6-(Co-Na) g
[NSP]@MWNTs h
Cr-MCM-41 i
3
8
1
12
8
9
1
10
11
1
24
10
PS-DA-Cu j
1
51.9
26

This
work
12
13
DIL-L/W(33).k
DIL-A/W(33).k
15
15
70
70
6
8
96
93
This
work
* In the presence of solvent (cyclohexane)
Reaction conditions:
a catalyst, 50 mg; cyclohexene, 5 mL; oxygen balloon
b catalyst, 50 mg; cyclohexene, 20 mL
c catalyst, 50 mg; cyclohexene, 10 mL
d catalyst, 200 mg; cyclohexene, 20 mL
e catalyst, 200 mg; cyclohexene, 20 mL
f catalyst, 0.004 mol Co; cyclohexene, 2 mmol
g catalyst, 1.5 mg; cyclohexene, 2 mL
h catalyst, 1.2 × 10 -5 mol; cyclohexene, 10 mL
i catalyst, 20 mg; cyclohexene, 1 g; glass reactor equipped with oxygen balloon
j catalyst, 2 mg; cyclohexene, 2 mL; glass reactor
k catalyst, 20 mg; cyclohexene, 2 mL, titanium batch reactor
4. CONCLUSION
On the basis of two dicationic ionic liquids and tungstophosphoric and
molybdophosphoric acids, different catalysts were prepared and characterized by FESEM-EDX,
XRD, TGA, BET and FT-IR techniques. The catalytic activity of the prepared catalysts with
molecular oxygen as oxidant for oxidation of cyclohexene was studied. The catalyst DIL-
L/W(33 wt%) was found to have a desirable catalytic activity for oxidation of cyclohexene. By
considering various parameters the reaction conditions were optimized. The maximum
conversion of cyclohexene was 96% and selectivity for the main product i.e., 2-cyclohexene-1-
one was 86%. Moreover, we have made efforts to elucidate the structure of DIL-L/W
nanocomposite and calculate the interaction energy of tungstophosphoric acid with DIL-L on the
basis of semi-empirical method.
Supporting Information
27

1
Additional experimental data including some FT-IR and H NMR spectra, TGA curve, and the
optimized structure of prepared catalysts
Acknowledgments
Thanks are due to the Research Council of Isfahan University of Technology for supporting
of this work.
References
[1] D. Yang, T. Jiang, T. Wu, P. Zhang, H. Han, B. Han, Catal. Sci. Tech. 6 (2016) 193-200.
[2] M.L. Kuznetsov, B.G.M. Rocha, A.J.L. Pombeiro, G.B. Shulpin, ACS Catal. 5 (2015) 3823-
3835.
[3] C. Yin, Z. Yang, B. Li, F. Zhang, J. Wang, E. Ou, Catal. Lett. 131 (2009) 440-443.
[4] X. Yun, X. Hu, Z. Jin, J. Hu, C. Yan, J. Yao, H. Li, J. Mol. Catal. A: Chem. 327 (2010) 25-
31.
[5] M. Shahabinejad, G. Ghasemi, M. Martínez-Huerta, M. Ghiaci, J. Mol. Catal. A: Chem. 406
(2015) 118-126.
[6] M. Ghiaci, N. Dorostkar, M.V. Martínez-Huerta, J. Fierro, P. Moshiri J. Mol. Catal. A:
Chem. 379 (2013) 340-349.
[7] X. Sang, J. Zhang, T. Wu, B. Zhang, X. Ma, L. Peng, B. Han, X. Kang, C. Liu, G. Yang,
RSC Adv. 5 (2015) 67168-67174.
[8] H. Shima, M. Tanaka, H. Imai, T. Yokoi, T. Tatsumi, J.N. Kondo, J. Phys. Chem. C 113
(2009) 21693-21699.
[9] A.A. Rosatella, C.A.M. Afonso, L.C. Branco, J. Chem. Educ. 88 (2011) 1002-1003.
[10] C.A. Bradley, M.J. McMurdo, T.D. Tilley, J. Phys. Chem. C 111 (2007) 17570-17579.
28

[11] S.N. Dhuri, K.B. Cho, Y.M. Lee, S.Y. Shin, J.H. Kim, D. Mandal, S. Shaik, W. Nam, J.
Am. Chem. Soc. 137 (2015) 8623-8632.
[12] P.J. Cordeiro, P. Guillo, C.S. Spanjers, J.W. Chang, M.I. Lipschutz, M.E. Fasulo, R.M.
Rioux, T.D. Tilley, ACS Catal. 3 (2013) 2269-2279.
[13] H.G.T. Nguyen, L. Mao, A.W. Peters, C.O. Audu, Z.J. Brown, O.K. Farha, J.T. Hupp, S.T.
Nguyen, Catal. Sci. Technol. 5 (2015) 4444-4451.
[14] D.R. Godhani, H.D. Nakum, D.K. Parmar, J.P. Mehta, N.C. Desai, J. Mol. Catal. A: Chem.
415 (2016) 37-55.
[15] A. Najafian, M. Rabbani, R. Rahimi, M. Deilamkamar, A. Maleki, Solid State Sci. 46
(2015) 7-13.
[16] R. De Paula, I.C.M.S. Santos, M.M.Q. Simões, M.G.P.M.S. Neves, J.A.S. Cavaleiro, J.
Mol. Catal. A: Chem. 404-405 (2015) 156-166.
[17] P. Cancino, V. Paredes-García, P. Aguirre, E. Spodine, Catal. Sci. Tech. 4 (2014) 2599-
2607.
[18] G. Li, Y. Ding, J. Wang, X. Wang, J. Suo, J. Mol. Catal. A: Chem. 262 (2007) 67-76.
[19] L. Rožić, B. Grbić, S. Petrović, N. Radić, L. Damjanović, Z. Vuković, Mater. Chem. Phys.
167 (2015) 42-48.
[20] M.N. Gorsd, M.N. Blanco, L.R. Pizzio, Mater. Chem. Phys. 171 (2016) 281-289.
[21] A.N. Kharat, S. Moosavikia, B.T. Jahromi, A. Badiei, J. Mol. Catal. A: Chem. 348 (2011)
14-19.
[22] Y. Leng, J. Wang, D. Zhu, L. Shen, P. Zhao, M. Zhang, Chem. Eng. J. 173 (2011) 620-626.
[23] H. Li, Y. Qiao, L. Hua, Z. Hou, B. Feng, Z. Pan, Y. Hu, X. Wang, X. Zhao, Y. Yu,
ChemCatChem. 2 (2010) 1165-1170.
29