Scheme 1 shows the sequence of steps in the preparation of the L-Co (II) -γ-Fe2O3 catalyst. In the first step, ionic liquid (A) was synthesized from the reaction between N-methyl imidazole and 3-bromopropyl amine hydrobromide. Then, the amine group of IL was reacted with one of the two carbonyl groups of 2, 6-diacetyl pyridine and produced B. In the next step in another Schiff-base reaction, the amine group of L-lysine amino acid was reacted with the carbonyl group of B and prepared L. Then, a certain molar ratio of L and CoCl2 was refluxed for 24 h to prepare the Co-complex. Finally, Co-complex was stabilized on γ-Fe2O3 NPs to prepare L-Co (II) -γ-Fe2O3.
1H-NMR, 13C-NMR, XRD, SEM, TEM, EDX, CHN, VSM, ICP-OES, and FT-IR analyses were employed to confirm the synthesized catalyst and ligand.
Identification of ligands and heterogeneous chiral catalyst
The FT-IR spectrum confirms all the steps of the ligand and catalyst preparation. In the spectrum (a) of Fig. 1, the vibrations appearing in the range of 2970–3100 cm-1 are related to the stretching vibrations of the C-H bond of the imidazole ring, and the vibrations of 2856 and 2933 cm-1 are related to the C-H bonds of the alkyl chain connected to the imidazole ring. The peaks at 1641 cm-1 and 1710 cm-1 are related to the C = N bond resulting from the Schiff base and the C = O bond of the remaining carbonyl group of the 2,6-diacetyl pyridine structure, respectively. Peaks related to the C-C and C-N bonds of the imidazole ring also appeared in 1176, 1462, and 1570 cm-1. In the spectrum (b) of Fig. 1, the peaks appearing at 1586 cm-1 and 1408 cm-1 are related to the vibrations of the COO- and the peaks in the region of 3100 cm-1 and 1631 cm-1 are related to the vibrations of the NH3+. The N-H vibrations of L-lysine in the region 2500–3300 cm-1 have appeared as a broadening peak, and the vibration of the area 3462 cm-1 is related to the hydroxyl group of L-lysine. Also, the disappearance of the carbonyl peak in the area of 1710 cm-1 compared to spectrum A indicates the formation of the Schiff open group of the amino group of L-lysine with the carbonyl group of the remaining compound of 2,6- DAP.
The effects of the cobalt complex formation with the ligand can be well seen in the (c) spectrum. The addition of the Co-N vibration in the region of 436 cm-1 and the shift of the N = C peak from 1631 to 1596 cm-1 confirm this well. In the d spectrum, the appearance of two peaks in the 584 cm-1 and 629 cm-1 areas is related to the Fe-O bond. Also, the disappearance of broad peaks in the regions of 2500–3300 cm-1 and 3400–3600 cm-1 and the presence of two peaks in the area of 3300–3600 cm-1 related to chiral amines indicate the connection of the complex from the acidic part to the magnetic substrate. The spectrum of (e) is related to the catalyst after five times of recovery and use. As is clear, it contains all the indicator peaks related to the catalyst (spectra d), which shows the stability of the catalyst in the reaction condition.
Scanning Electron Microscopy (SEM) examined the catalyst's surface morphology, and the images showed that the catalyst has an irregular and agglomerated morphology. The reason for this morphology can be the magnetic properties of the catalyst, which caused the magnetic particles to stick together (Fig. 2).
The magnetism of γ-Fe2O3 NPs is an excellent property for use in catalysts because of their easy separation. The magnetic properties of γ-Fe2O3 NP NPs, catalyst, and reused catalyst were investigated by VSM analysis at room temperature. As the curves in Fig. 3 show, the γ-Fe2O3 NPs have a saturation magnetization of about 49.264 emu/g, while the catalyst and reused catalyst show values of 42.236 emu/g and 37.758 emu/g for magnetism, respectively. This reduction in magnetic property is mainly due to the catalyst being grafted to the nanoparticles and creating a layer on them. It should be noted that no magnetic hysteresis was observed in any of the samples. Comparing the reused catalyst with the fresh catalyst shows it has maintained its magnetic properties well after use. This superparamagnetic behavior causes the catalyst to be separated from the reaction mixture by an external magnetic field.
Thermo gravimetric analysis (TGA) investigated the catalyst's thermal stability. As can be seen in Fig. 4, with the increase in the decomposition temperature from 20 to 600°C, two decreases in the weight of the catalyst are observed. The first decrease (4%) that happened below 200°C is related to the removal of water molecules absorbed by the catalyst, and the main decrease of 14% in the temperature range of 200 to 500°C is related to the separation of the ionic liquid group from the catalyst. Considering that epoxidation reactions are carried out at temperatures below 100°C, it can be concluded that the catalyst has good stability.
EDX analysis data for magnetic chiral catalyst is shown in Fig. 5. All the constituent elements of the catalyst are well seen in the pattern. As can be seen, the catalyst contains carbon, nitrogen, oxygen, iron, and cobalt, which account for 11.2, 6.21, 31.92, 46.29, and 4.37% of the weight of the catalyst, respectively.
The amount of cobalt metal in the L-Co (II) - γ-Fe2O3 catalyst was measured by ICP analysis. Analysis showed that 8.73 mg of cobalt (in 100 mg of catalyst) was loaded into the catalyst.
The XRD pattern obtained from the final structure of the catalyst is shown in Fig. 6. In the pattern, the characteristic peaks of diffraction in 2 θ = 30°, 35°, 43°, 54°, 57°, 63°, 71°, and 45° corresponded to the indexed (220), (311), (400), (422), (511), (440), (620) and (533), respectively. The pattern obtained was in good agreement with the standard card... and confirmed the structure of γ-Fe2O3.
Study of the catalytic property
After preparing and characterization of the L-Co (II)-γ-Fe2O3 catalyst, to investigate it in the epoxidation of alkenes and to optimize the reaction conditions, first, the epoxidation of styrene (General method for epoxidation of alkenes) was performed in the presence of the catalyst. After confirming the catalytic properties of the catalyst, the reaction conditions were optimized. The reaction was first performed for optimization in the presence of air, H2O2, O2, TBHP, and KHSO5 (Oxone). According to the obtained results, TBHP had the best efficiency in producing epoxide products and was chosen as the optimal oxidant (Table 1, rows 1–5). The styrene epoxidation reaction was carried out in the presence of different solvents and free solvents. Acetonitrile and ethyl acetate showed almost similar efficiency; ethyl acetate was chosen as the superior solvent due to its Green and environmentally friendly properties (Table 1, entries 5–12). The reaction in solvent-free conditions had a meager efficiency after 4 hours (Table 1, rows 12). The reaction was investigated in the presence of different amounts of catalyst and without catalyst. The results showed that the amount of 2% molar catalyst has the highest efficiency, and in the absence of the catalyst, the reaction did not show any progress after 4 hours (Table 1, rows 13–18). Finally, the reaction was checked at different temperatures, and the highest efficiency was observed at 80°C (Table 1, rows 19–24). According to the results obtained in Table 1, the optimal conditions of the reaction were chosen: TBHP as an oxidant, ethyl acetate as a solvent, the amount of 2 mole % of the catalyst, and a temperature of 80°C.
Table 1
Optimization of parameters affecting styrene epoxidation
Entry a
|
Oxidant
|
Solvent
|
Catalyst (mol %)b
|
Temperature (°C)
|
Conversion (%)c
|
1
|
Air
|
CH3CN
|
3
|
70
|
18
|
2
|
O2
|
CH3CN
|
3
|
70
|
55
|
3
|
H2O2
|
CH3CN
|
3
|
70
|
82
|
4
|
TBHP
|
CH3CN
|
3
|
70
|
95
|
5
|
Oxone
|
CH3CN
|
3
|
70
|
81
|
6
|
TBHP
|
H2O
|
3
|
70
|
48
|
7
|
TBHP
|
EtOH
|
3
|
70
|
62
|
8
|
TBHP
|
MeOH
|
3
|
70
|
66
|
9
|
TBHP
|
EtOAc
|
3
|
70
|
90
|
10
|
TBHP
|
DCM
|
3
|
70
|
28
|
11
|
TBHP
|
CH3Cl
|
3
|
70
|
33
|
12
|
TBHP
|
Free
|
3
|
70
|
trace
|
13
|
TBHP
|
EtOAc
|
0.5
|
70
|
81
|
14
|
TBHP
|
EtOAc
|
1
|
70
|
87
|
15
|
TBHP
|
EtOAc
|
2
|
70
|
95
|
16
|
TBHP
|
EtOAc
|
4
|
70
|
90
|
17
|
TBHP
|
EtOAc
|
6
|
70
|
78
|
18
|
TBHP
|
EtOAc
|
8
|
70
|
60
|
19
|
TBHP
|
EtOAc
|
Free
|
70
|
0
|
20
|
TBHP
|
EtOAc
|
2
|
Rt
|
42
|
21
|
TBHP
|
EtOAc
|
2
|
50
|
62
|
22
|
TBHP
|
EtOAc
|
2
|
60
|
89
|
23
|
TBHP
|
EtOAc
|
2
|
80
|
98
|
24
|
TBHP
|
EtOAc
|
2
|
90
|
94
|
.a reaction condition: styrene (1 mmol), Oxidant (1mmol), and 4 h reactions time |
b based on Co metal measured by ICP |
c based on styrene and measured by GC |
Catalytic epoxidation of alkenes
We investigated the efficiency of the L-Co (II) -γ-Fe2O3 catalyst under optimal conditions obtained for the epoxidation of different alkenes. 1 mmol of alkene was subjected to an epoxidation reaction in the presence of 1 mmol of TBHP under optimal conditions for three hours. The results are shown in Table 2. The amount of conversion to product well confirms that the presence of steric crowding causes a decrease in conversion because it becomes difficult for the catalyst to approach the double bond (Table 2, comparing entries 1 with 2, as well as entries 6 and 7 with 8 and 9). It was found that C = C conjugation leads to a decrease in efficiency (Table 2, entries 4, 6, and 7). Evaluation of epoxide production yield showed that the catalyst performed very well in all reactants. Also, the investigation of the enantiomer excess showed that due to catalyst bulkiness, it worked selective in compounds with steric effects and produced more of a specific type of enantiomer (Table 2, entries 2, 5, 6, and 7). This issue can be observed well in compounds with a small steric hindrance around the C = C, such as linear alkenes (Table 2, entries 8 and 9).
Table 2. Epoxidation of various alkenes catalyzed by L-Co (II) -γ-Fe2O3 catalyst
a Reaction conditions: substrate (1 mmol), EtOAc (3 ml), oxidant (1 mmol), L-Co (II) -γ-Fe2O3 catalyst (2 mol% based on Co), and 4 h reactions time.
b Determined by GC with a Shimadzu CBP5 column (30 m× 0.32 mm × 0.25 mm).
c Performed by GC with a Agilent HP column (19091G-B213, 30 m × 0.25 mm × 0.25 µm).
d %ee = [A (major enantiomer) –A (minor enantiomer)] × 100/[A(major enantiomer) + A(minor enantiomer)].
Recycling and Leaching Test of Catalyst
One of the advantages of heterogeneous catalysts is the ease of reuse, high stability, and low leaching. The stability and reusability of catalyst 1 in the styrene epoxidation reaction were studied under optimal conditions. The catalyst was used in the reaction. After 4 hours, it was separated from the reaction by an external magnet, washed 3×10 ml with EtOH, dried in the oven without any purification, and used again. The results obtained for 5 consecutive uses of the catalyst are given in Fig. 7. The results showed that the conversion rate and yield during 5 times recycling, decreased by only 5% and 3% respectively, indicating the high stability of the catalyst and its very little leaching. Also, the ICP analysis of the catalyst after 5 times of recycling showed 8.55 mg cobalt in the catalyst, which showed 2% leaching of cobalt compared to the unused catalyst.
It was used in a styrene epoxidation reaction to confirm that the catalyst was not leached in the reactions. After 2 hours, it was separated from the reaction by an external magnet, the reaction had 42% conversion at the time of separation, and after 4 hours of reaction time, no improvement in the conversion rate was observed. These results and the results of VSM analysis for the catalyst after reuse (Fig. 4, green graph) confirmed the stability and recyclability of catalyst 1 as an efficient catalyst for the epoxidation of alkenes.
Mechanism study of catalytic epoxidation of alkenes
The proposed mechanism for the epoxidation of alkenes in the presence of a Schiff-base catalyst is presented in Schematic 3. In this mechanism, first, TBHP is coordinated to the cobalt complex, one chlorine is released in the form of HCl and the peroxo-metal compound B is formed. Then, the alkene enters the catalytic site, and takes an oxygen atom from the peroxo-metal compound to form the epoxide, while another chlorine atom is removed, leading to the formation of compound D. Finally, in the presence of HCl and TBHP, compound D is converted back to compound B, completing the catalytic cycle.
Comparison of our catalyst with other reported catalysts
To examine the strengths and weaknesses of the catalyst, this work was compared with previous catalysts reported in the literature for styrene epoxidation. Styrene oxide was prepared in the presence of L-Co (II) -γ-Fe2O3 catalyst in a short time at 80°C in EtOAc. The advantages of this catalyst include short time, high stability, easy separation, high selectivity, and most importantly, its excellent performance in ethyl acetate solvent as a green solvent. As can be seen in Table 3, this method has the highest efficiency in less time than other works, the reason can be attributed to the presence of ionic liquid groups in the structure of the catalyst, which helps to increase the solubility of the catalyst in the polar environment of the reaction, and also It causes its bulkiness and high selectivity.
Table 3
Comparison catalyst activity of L-Co (II) -γ-Fe2O3 catalyst whit other catalysts reported for styrene epoxidation
Entry
|
Catalyst
|
Conditions
|
Time (h)
|
Epoxide
Yield (%)
|
References
|
1
|
FeL4 -SBA
|
CH3CN/isobutyraldehyde/air/ 80°C
|
8
|
80
|
[44]
|
2
|
Cobalt(III) Schiff base complex
|
CH3CN/isobutyraldehyde/air/ 60°C
|
8
|
92
|
[45]
|
3
|
GO/Fe3O4@PAA-Co(II)
|
CH2Cl2 / PhIO / Rt
|
6
|
85
|
[46]
|
4
|
CoFe2O4 nanoparticles
|
CH3CN / TBHP/ 80°C
|
9
|
82
|
[47]
|
5
|
Co-GO
|
CH3CN/ air/ isobutyraldehyde / 80°C
|
8
|
59.5
|
[48]
|
6
|
Co–Salen–GO
|
CH3CN/air/ isobutyraldehyde/ 80°C
|
6
|
70
|
[49]
|
7
|
L-Co (II) -γ-Fe2O3
|
EtOAc/TBHP/80°C
|
4
|
98
|
Present study
|