Introduction

The catalytic oxidation of organic materials is a substantial process in organic synthesis in order to producing practical oxidation products [1,2,3,4] including epoxides, which are useful synthetic intermediates and because of their high reactivity have attracted a great deal of attention for decades due to their use in a broad range of synthesis reactions [3]. They can be applied for the production of plasticizers, cosmetics, epoxy resins, surfactants, paints, perfumes, etc. [5, 6]. The epoxidation of alkenes in the presence of an oxygen source is a well-known fundamental route for the preparation of these compounds [7,8,9,10]. Sulfoxides are another class of important products in the oxidation of sulfides [11,12,13]. Besides their inevitable role in organic synthesis, sulfoxides have a lot of applications in biological and medicinal such as: antibacterial, antifungal, anti-hypertensive, anti-ulcer and anti-atherosclerotic [14,15,16].

Various heterogeneous and homogeneous catalysts have been applied for oxidation and epoxidation of organic compounds. Although homogeneous catalysts are more efficient than heterogeneous types, they suffer from some problems concerning to tedious separation and recovery of the catalyst from the reaction mixture [17, 18]. From this point of view, in order to solving this obstacle, numerous strategies have been developed for immobilization of the complexes on a suitable solid support [19] such as sepiolites [20], silica [21], zeolite [22], carbon based material [23], polymer [24], and nanomaterials [25].

The application of NPs have been manifested in various fields from chemistry, agriculture, to medicine and pharmacy [26, 27]. In chemistry, due to their high aspect ratio (surface to volume ratio) of the nanoparticles, they have become a powerful tool as supporting agent; also, their catalytic activities have been proven for decades [28, 29]. In particular, the use of magnetic NPs is a smart strategy in order to design a heterogeneous catalyst; because they not only utilize magnetically recoverability arising from super paramagnetism property, but also they are safe and biocompatible compounds that could be easily functionalized with a broad range of compounds to nailing a special application [30]. Motivated by this concept, tremendous attempts have been made for their applications as catalyst [8]. In this viewpoint, various methods have been reported for sulfide oxidation that come as follow: (1) Fe3O4@SiO2-tris(8quinolinolato)iron [31], (2) Fe3O4@tryptophan-Ni(II) complex [19], (3) Fe3O4@SiO2@DOPisatin-M(II) (M=Ni, Cu) [32], (4) Fe3O4@Cu–S-(propyl)-2-aminobenzothioate [33], (5) Fe3O4@SiO2-APTES-FeLGDC [34], (6) Fe3O4–TBA-Br3 [30], (7) MoO2-CML–Fe3O4@SiO2 [11], (8) Fe3O4@Gly@r-Py@VO3 [35], (9) M/isatin@Fe3O4 (M=Zr(IV), Cu(II)) [36], (10) Fe3O4@PEI@Br [37] and (11) Fe3O4@CP(AEAM)/W [38]. Recently, Rayati et al., reported a Mn-Schiff base complex supported on magnetic nanoparticles for sulfide and alkene oxidation [14]. Also, different methodologies based on magnetic nanoparticles have been developed for the epoxidation of olefins. For example: MoO3(sal-phz)/Fe3O4 [39], MnFe2O4@Mo(Vi) Schiff base complex [40, 41], RGO/Fe3O4@C-Salen-MoO2 [42], Fe3O4@SiO2@APTMS@Co(II) complex [43], γ-Fe2O3TBHP [44], Fe3O4/tart/Mn(TCPP)Cl [8], Fe3O4@SiO2@APTMS@V(v) complex [28], Fe3O4@Cr(VI) hybrid [26] and Fe3O4Mn(III) porphyrin [45] were some of the recent advances nanomagnetite based catalyzed methods regarding epoxidation of alkenes.

However, most of the reported catalytic systems utilize the transition metal complexes supported on the surfaces of nanoparticles [43]. The instability of some heterogeneous complexes, metal leaching concerns, use of toxic heavy metals, metal oxidation during the process, high cost, tedious procedure, and et cetera, are some disadvantages regarding to the complexes supported on NPs. So, finding a more sustainable, cost effective and ecofriendly methodology for efficient oxidation of organic compounds is in demand.

We previously reported the epoxidation of alkenes by Fe3O4@SiO2 nanoparticles [46]. In this paper, we introduce FeNi3@SiO2 magnetic nanoparticles, as a heterogeneous recoverable catalyst, for efficient epoxidation of alkenes as well as oxidation of sulfides in the presence of meta-chloroperoxybenzoic acid (m-CPBA) at room temperature.

Experimental

Material and instruments

Tetraethoxysilane (TEOS), NiCl2·6H2O, FeCl2·4H2O, polyethylene glycol (average Mn = 6000), NaOH, N2H4 hydrate (80 wt%), ammonia solution (25 wt%) and Na2SO4 were purchased from Sigma Aldrich and Merck and used without purification. All alkenes and sulfides including: styrene, α-methyl styrene, 1H-indene, cyclohexene, 4-chlorostyrene, cyclooctene, 1-hexen, cis- and trans-stilbene, thioanisole, 2-chlorothioanisole, 4-methoxythioanisole, 4-methylthioanisole, 4-nitrothioanisole, 4-bromothioanisole, diphenylsulfane, and dibutylsulfane were of the highest grade available (Aldrich) and were used as received. Dichloromethane (DCM, anhydrous 99.8%), ethyl acetate (EtOAC), ethanol (EtOH, absolute grade. ≥ 99.8%), CH3CN, CHCl3, THF, m-CPBA, NaIO4, t-BOOH, PhI(OAC)2, PhIO, H2O2 (30 wt%) and Oxone was purchased from Sigma Aldrich and Merck and used without further purification. Fourier transform infrared (FTIR) spectra were recorded on a JASCO FT/IR 4600 instrument in the range of 450–4000 cm−1. The crystal structure of the FeNi3 and FeNi3@SiO2 NPs was studied by a Bruker AXS D8-advance X-ray diffractometer using Cu-Kα radiation. Transmission electron microscopy (TEM) images were obtained on a JEOL-2010 transmission electron microscope with an accelerating voltage of 200 kV. The magnetization behavior of the NPs were investigated on a Lake Shore vibrating sample magnetometer (VSM) at room temperature. Progress of the reactions and reaction conversions were conducted on a YL 6100 gas chromatograph system (GC) with a CBP5 column (Shimadzu 30 m × 0.32 mm × 0.25 mm). Conversion and selectivity of the products were measured qualitatively by gas chromatography. Conversion of the product was obtained by comparing the area of the product and the starting material. For each experiment, the resulting mixture (0.2 μL) was injected to GC instrument then the selectivity of the oxidation product was calculated by the following Eq. (1) [6]:

$${\text{Epoxide/sulfoxide selectivity = }}\frac{\text{GC peak area of epoxide/ sulfoxide}}{\text{GC peak area of all products}} \times 100.$$
(1)

Preparation of FeNi3 NPs

FeNi3 NPs were prepared using a wet chemical approach according to the previously reported work [46, 47]. Briefly, FeCl2·4H2O (10 mmol) and NiCl2·6H2O (30 mmol) were dissolved into 200 mL distilled water. Then, 0.1 g of polyethylene glycol (average Mn 6000), as a surfactant agent, was added to the mixture. NaOH (10%, w/w) was added dropwise until the pH of the solution reached to 12, followed by the addition of N2H4·H2O 80% (N2H4·H2O/FeNi3 molar ratio was 24:1). The reaction was stirred for 24 h at room temperature under nitrogen atmosphere. During the reaction, the pH value was controlled and adjusted in the range of 12–13. After completion of the reaction, the magnetic nanoparticles FeNi3 NPs were collected with an external magnetic field and washed with deionized water (5 × 10 mL) and dried at oven (50 °C) for 12 h.

Preparation of FeNi3@SiO2 core–shell NPs

FeNi3@SiO2 core–shell NPs were prepared by a modified Stöber method [25]. FeNi3 NPs (0.5 g) were dispersed in a mixture of ethanol and deionized water (80:20 mL) and concentrated ammonia aqueous solution (2.0 mL, 25 wt%). Tetraethylzorthosilicate (0.20 g) was added to the above mixture and the mixture vigorously stirred at room temperature for 24 h. Then, FeNi3 NPs were magnetically separated, rinsed with deionized water (5 × 10 mL), and dried at 60 °C.

General procedure for epoxidation of alkenes and oxidation of sulfides catalyzed by FeNi3@SiO2

The oxidation reactions were performed in the presence of 50 mg and 40 mg of the catalyst respectively, for epoxidations and sulfide oxidation. Typically, a mixture of the catalyst dispersed in dichloromethane (3.0 mL), was sonicated for 10 min at room temperature. The substrate (1.0 mmol) and m-CPBA (2.0 mmol) were added to the above flask containing the catalyst and oxidant. The mixture was stirred at room temperature for appropriate time. The progress of the reaction was screened using GC or TLC. Upon the reaction completion, the catalyst was separated by an external magnet, washed with EtOH (2 × 10 mL), dried at 50 °C in air atmosphere and set aside for the next run. The remaining solution was washed with NaOH (1 N, 8.0 mL) and brine (8.0 mL), then dried over Na2SO4. Then, the solvent was removed using rotary evaporator. Conversion and selectivity of the products were determined by GC.

Results and discussion

Characterizations of the catalyst

The FTIR spectra of FeNi3 and FeNi3@SiO2 were shown in Fig. 1. In Fig. 1a, the absorption bands at 560–590 and 3400 cm−1 were assigned to stretching vibrations of Fe–O and O–H respectively. Three main characteristic peaks at 560–590, 954, and 3400 cm−1, showing the stretching vibrations of Fe–O, Si–OH, and O–H. Also, the bands at 756 and 1105 cm−1 are due to symmetric and asymmetric stretching vibrations of Si–O–Si. The results were completely in agreement with the reported FTIR spectra for coating of silica shell on the surface of NPs [25, 47].

Fig. 1
figure 1

FTIR spectra of a FeNi3 and b FeNi3@SiO2

Full size image

X-ray diffraction (XRD) pattern of FeNi3 and FeNi3@SiO2 NPs are shown in Fig. 2. FeNi3 NPs exhibited good crystalline structure, which were in agreement with the standard FeNi3 (JCPDS Card No. 19-0629) [46]. Three characteristic peaks demonstrated at 2θ = 44.3°, 51.5°, 75.9°, which were marked based on their related indices (111), (200) and (220) planes, are obtained for FeNi3 (Fig. 2a). The presence of these peaks in XRD spectrum of FeNi3@SiO2 proved that the surface modification of FeNi3 do not lead to their phase change. An amorphous peak appeared at 2θ = 13.2° was assigned to silica supported on FeNi3 (Fig. 2b).

Fig. 2
figure 2

XRD pattern of FeNi3 (a) and FeNi3@SiO2 (b)

Full size image

Fig. 3 shows TEM image of FeNi3@SiO2 MNPs. The nanoparticles have spherical shape with average size of 25 nm. Coated silica was discriminated by a different phase with average diameter of 2 nm, confirming the favored coating of FeNi3 by silica layers (Fig. 3).

Fig. 3
figure 3

TEM image of FeNi3@SiO2

Full size image

The magnetic behavior of the NPs was investigated by VSM analysis (Fig. 4). According to the Fig. 4a, FeNi3 and FeNi3@SiO2 did not show hysteresis phenomenon and have zero coercivity at room temperature. This behavior belongs to superparamagnetic nanoparticles in agreement with the literature [30]. The saturation magnetization values for FeNi3 and FeNi3@SiO2 NPs were found at 61.648 and 44.31 emu/g, respectively. Although the magnetization of FeNi3 considerably reduced, FeNi3@SiO2 NPs can still be efficiently separated by using a simple external magnet from the solution.

Fig. 4
figure 4

VSM curves of a FeNi3 and b FeNi3@SiO2

Full size image

Optimization of the reaction parameters

First, in order to find the best conditions for the FeNi3@SiO2-catalyzed epoxidation of alkenes and the oxidation of sulfide, styrene and methyl phenyl sulfide were used as the model substrates for epoxidation of alkene and oxidation of sulfide.

Catalyst amount

In order to finding optimum amount of the catalyst, the model reactions were investigated by varying amount of the catalyst (Fig. 5). The experiments elucidated that 40 mg and 50 mg were sufficient enough for achievement of the most efficiency for the methyl phenyl sulfoxide and styrene oxide, respectively. No more efficiency was attained for the two model reactions with higher aforementioned above amount of the catalyst (Fig. 5). The reactions in the absence of the catalyst (blank experiments) gave just 10% conversion for both substrates (Fig. 5).

Fig. 5
figure 5

Optimization of the catalyst amount for epoxidation of styrene and oxidation of methyl phenyl sulfide. Reaction conditions for the two oxidations: styrene or methyl phenyl sulfide (1.0 mmol), m-CPBA (2.0 mmol), DCM (3.0 mL), FeNi3@SiO2 (Catalyst), 1.5 and 1.0 h for oxidation of styrene and methyl phenyl sulfide oxidation respectively, r.t. ★ GC analysis

Full size image

Effect of oxidant

Various types of oxidant were tested for the model reactions (Fig. 6). It was observed that m-CPBA gives the most possible conversion of epoxy styrene (97%) and methyl phenyl sulfoxide (99%) than other tested oxidants. Moreover, it was found that the highest conversion was obtained with 2.0 equivalent of the m-CPBA in the presence of the FeNi3@SiO2. Low conversion of the products (≤ 35%) were obtained with H2O2, UHP, PhI(OAc) and PhIO. On the other hand, t-BuOOH and oxone gave a moderate yield (Fig. 6). In a blank experiment, the reactions were done in FeNi3 and/or FeNi3/SiO2 NPs alone (Not shown in the figure). The results did not show any reaction progress and they were inactive towards the epoxidation of styrene or oxidation of methyl phenyl sulfide and demonstrated the mandatory role of oxidant as oxygen source.

Fig. 6
figure 6

Influence of oxidant type on epoxidation of styrene and oxidation of methyl phenyl sulfide. Reaction conditions for the two oxidations: styrene or methyl phenyl sulfide (1.0 mmol), oxidant (2.0 mmol), DCM (3.0 mL), FeNi3@SiO2 (Catalyst, 40 and 50 mg for oxidation of methyl phenyl sulfide and styrene, respectively), 1.5 and 1.0 h for oxidation of styrene and methyl phenyl sulfide oxidation respectively, r.t. ★ GC analysis. ★★ KHSO5. ★★★ Urea-Hydrogen peroxide

Full size image

Effect of solvent

The two model reactions were checked in different solvents including EtOAc, CH3CN, H-2O, EtOH, H2O: EtOH (1:1 v/v), CHCl3, THF and DCM (Fig. 7). As shown in Fig. 7, H2O and EtOH give the lowest reaction conversions. EtOAc (75% epoxy styrene, 79% methyl phenyl sulfoxide), CH3CN (65% epoxy styrene, 65% methyl phenyl sulfoxide) and CHCl3 (55% epoxy styrene, 64% methyl phenyl sulfoxide) provided moderate efficiency for the both oxidation systems. The highest conversion was obtained in DCM (97% epoxy styrene, 99% methyl phenyl sulfoxide) and provide the oxidation products in quantitative yield (Fig. 7).

Fig. 7
figure 7

Effect of solvent on epoxidation of styrene and oxidation of methyl phenyl sulfide. Reaction conditions for the two oxidations: styrene or methyl phenyl sulfide (1.0 mmol), m-CPBA (2.0 mmol), solvent (3.0 mL), FeNi3@SiO2 (Catalyst, 40 and 50 mg for oxidation of methyl phenyl sulfide and styrene respectively), 1.5 and 1 h for oxidation of styrene and methyl phenyl sulfide oxidation respectively, r.t. ★ GC analysis. ★★ 1:1. The conversion in solvent free conditions was 8% for the two experiments (not shown in the figure)

Full size image

Effect of temperature

The effect of temperature was also investigated for the reactions. The model reactions were performed at 0 and 40 °C and under reflux conditions. The oxidation reactions exhibited the same behavior: the reaction rates increased with temperature. No change in reaction selectivity was observed at 0 °C, but the selectivities of the both systems slowly decreased with temperature (Fig. 8). Due to economical aspects, room temperature was chosen in this study with a very slightly difference from reflux conditions (Fig. 8).

Fig. 8
figure 8

Influence of temperature on styrene epoxidation (grey columns and markers) and oxidation of methyl phenyl sulfide (black columns and markers)

Full size image

The selectivity of the reactions were investigated by GC. Fig. 9 shows the results for the model reactions. As shown in the figure, epoxy styrene and methyl phenyl sulfoxide were obtained with excellent selectivity without observation of any other oxidation product demonstrating the selective performance of the catalyst. No sulfone was seen in the case of sulfide oxidation resulting further oxidation of sulfoxide (Fig. 9). This excellent selectivity was also obtained for all of substrates under the present study.

Fig. 9
figure 9

GC diagrams for a epoxidation of styrene and b oxidation of methyl phenyl sulfide catalyzed by FeNi3 in the presence of m-CPBA at room temperature

Full size image

Catalytic activity

With the optimum conditions in hand, i.e. substrate (1.0 mmol), DCM (3.0 mL), m-CPBA (2.0 mmol) and catalyst (40 mg for sulfide oxidation and 50 mg for epoxidation), room temperature, we applied the catalyst toward epoxidation of a variety of alkenes as well as oxidation of various sulfides to sulfoxides. A series of alkenes investigated are reported in Table 1. Terminal benzylic alkenes show the highest efficiency [Table 1, entry 1 (97%), entry 2 (97%), entry 5 (95)]. As expected [40], Cl as deactivation group for benzene ring, delayed the epoxidation and reduced the conversion (Table 1, entry 5). Lower efficiencies were achieved for internal alkenes, cis-stilbene, cyclohexene and cyclooctene, with 90% yield. Excellent selectivity was obtained for the all of substrates (GC analysis, 98–99%).

Table 1 Epoxidation of different alkenes catalyzed by FeNi3/SiO2 NPsa
Full size table

Also, the catalyst shows good activity toward sulfide oxidation. A series of sulfides were oxidized to sulfoxide by magnetic FeNi3@SiO2 NPs at room temperature. The results were tabulated in Table 2. The catalyst demonstrated high to excellent selectivity and reaction conversion as same as epoxidation reaction with average conversion of 97%.

Table 2 Preparation of sulfoxide derivatives using FeNi3/SiO2 catalyst
Full size table

In order to validate th recyclability of the catalyst, the epoxidation of styrene was chosen as model reaction. The catalyst was magnetically separated, washed with ethanol and deionized water, dried, and re-used directly for the next reaction run without any further purification. The residue mixture was analyzed by GC on every run. The results, as shown in Fig. 10, indicated insignificant reactivity loss after five runs, in which only 2% reduction in efficiency was seen on average per cycle. It is worth nothing that this method is superior to other methods containing transition metal complexes with metal leaching concern. Moreover, the resulting TEM image from the recovered catalyst after the 5th run showed that the NPs retained their spherical shape, although a slight agglomeration occurred. Also, the size of the NPs increased after 5th cycle and that could be responsible for decreasing in the conversion. This behavior is expected due to their high aspect ratio of the nanoparticles [25].

Fig. 10
figure 10

a Recyclability study of the catalyst on epoxidation of styrene. Reaction conditions: styrene (1.0 mmol), m-CPBA (2.0 mmol), DCM (3.0 mL), the catalyst (50 mg), 1.5 h, room temperature. b TEM image of the recovered FeNi3@SiO2 after the 5th run

Full size image

The literature review shows that hydroxyl groups on nano silicate are well able to form hydrogen bonds with carboxylic groups [48, 49] (ESI, Figs. 1, 2, 3). Thus, an H-bonding can be found between the surface of nanoparticles containing Si–OH groups and the hydroxyl/carbonyl groups of m-CPBA. The proposed mechanism for this system was shown in Fig. 11, for an epoxidation reaction catalyzed by FeNi3@SiO2 in the presence of m-CPBA. In this proposed mechanism, given the presence of silica supported on the FeNi3 nanoparticles (Fig. 11), m-CPBA was adsorbed on the catalyst surface by hydrogen bonding. Then, the alkene epoxidation with m-CPBA [50] can be carried out in the presence of FeNi3 NPs. Because of the high surface to volume ratio of the nanoparticles, a high surface area is provided for the oxidation reaction. So the reaction could be proceeded with high efficiency and selectivity on such huge surface. The same explanation could be introduced for sulfide oxidation to sulfoxide. However, the mechanism of such systems still requires future investigations.

Fig. 11
figure 11

A plausible reaction mechanism for catalytic epoxidation of alkene by FeNi3@SiO2 in the presence of m-CPBA

Full size image

Conclusions

In conclusion, a facile, general, cost effective and environmentally benign method have been developed for oxidation of sulfides and epoxidation of alkenes by fabrication of heterogeneous magnetically recoverable FeNi3@SiO2 nanoparticles as an efficient catalyst in the presence of m-CPBA at room temperature. The catalyst shows high selectivity toward sulfoxide and epoxide products, which no sulfone or other by-products were found in the mixture. The catalyst was characterized by FTIR, TEM, VSM and XRD analyses. Oxidation was take place on the surface of the FeNi3@SiO2 nanoparticles with a high accessible surface. The catalyst showed good versatility toward a broad spectrum of alkenes and sulfide for oxidation, where high to excellent reaction conversions were obtained for all of substrates studied in this paper. High stability, easy preparation, magnetically recoverable, simple handling, good catalytic activity are some highlighted properties of the catalyst that make this methodology as a promising tool for a broad scope of oxidation goals. Finally, the magnetic catalyst was simply recovered and reused for five consecutive runs without remarkable reactivity loss.