Iron oxide encapsulated by copper-apatite: an efficient magnetic nanocatalyst for N-arylation of imidazole with boronic acid

N-Arylation of imidazole was carried out with various arylboronic acids on iron oxide encapsulated by copper-apatite (Fe3O4@Cu-apatite), producing excellent yields. Firstly, the iron nanoparticles were prepared using a solvothermal method, and then they were encapsulated by copper-apatite to obtain magnetic Fe3O4@Cu-apatite nanocatalysts. Several physico-chemical analysis techniques were used to characterize the prepared nanostructured Fe3O4@Cu-apatite catalyst. The prepared Fe3O4@Cu-apatite was used as a nanocatalyst for N-arylation of imidazole with a series of arylboronic acids with different substituents to reaffirm the effectiveness of this magnetic nanocatalyst. The Fe3O4@Cu-apatite nanocatalyst can also be easily separated from the reaction mixture using an external magnet. More importantly, the as-prepared Fe3O4@Cu-apatite exhibited good reusability and stability properties in successive cycles. However, there was a notable loss of its catalytic activity after multiple cycles.


Introduction
Over the years, N-arylheterocycles have gained prominence due to their presence in a wide variety of natural products and bioactive compounds. They have also played an important role as building blocks in organic synthesis. 1,2 There is a special emphasis on N-aryl imidazole because it exhibits antipsychotic, antiallergic, and herbicidal 3,4 properties, as well as others. The development of a mild and highly efficient method for the synthesis of N-arylheterocycles over classical Ullman type, 5,6 nucleophilic aromatic substitution reactions, 7 or coupling with organometallic reagents has recently gained considerable attention in synthetic chemistry. 8,9 But the harsh conditions of these reactions, such as very high temperature and strong bases have restricted and limited their applications. At the turn of the 21 st century, recent development of Narylation with boronic acids unleashed the power of Cheteroatom bond formation reaction due to the mild reaction conditions (room temperature, weak base and ambient atmosphere). [10][11][12][13] However, the synthetic scope of this reaction is strongly limited due to some disadvantages, such as product contaminationtoxic waste produced aer separation of the catalyst, which cannot be easily recovered aer the reaction. 14 These disadvantages can be overcome by anchoring the metal on suitable supports, which can be easily recovered, and then potentially be reusable with a minimal amount of product contamination. Recently, diverse forms of heterogeneous catalysts have been developed for N-arylation, such as cellulose supported copper(0), 15 polymer supported Cu(II), 16 MCM-41-immobilized bidentate nitrogen copper(I), 17 CuO nanoparticles 18 and Cu-exchanged uorapatite. 19 Magnetic separation has also received a lot of attention as a solid catalyst separation technology, since it can be very efficient and fast. Numerous studies have focused on the immobilization of copper catalytic systems on a magnetic medium based on iron in order to separate them by the simple application of a magnet. 20,21 In the same way, Alper and co-workers have immobilized Cu(I) catalyzed on the surface of Fe 3 O 4 magnetic nanoparticle-supported L-proline as a recyclable and recoverable catalyst for N-arylation of heterocycles.
Calcium phosphate, such as hydroxyapatite and its derivates appear very attractive due to their ion-exchange capability, adsorption capacity, and acid-base properties. 22 Hydroxyapatite based materials have been used in the biomedical eld 23,24 lately and in some important chemical transformations. [25][26][27][28] The latter applications are related to its well-known ability to immobilize divalent and trivalent metal ions, by partial cationic exchange with calcium. [29][30][31][32][33] Recently, the possibility to combine the properties of apatite with other inorganic phases within coreshell nanostructures has attracted great attention. For instance, hydroxyapatite/TiO 2 , 34 hydroxyapatite/carbon nanotubes, 35 and hydroxyapatite/SiO 2 (ref. 36) core-shell nanoparticles have been studied for different applications, such as photocatalytic processes, uorescence imaging and drug delivery. Indeed, the combination of hydroxyapatite with iron oxide particles could be useful to design sorbents that combine large activity and easy recovery via magnetic separation. 37,38 Hence, several iron oxidehydroxyapatite nanocomposites have been described in the literature. [39][40][41][42][43][44][45] However, the Fe 3 O 4 @Cu-apatite has only rarely been explored, and to the best of our knowledge, the use of Fe 3 O 4 @Cu-apatite as a heterogeneous catalyst for the N-arylation of imidazole has not been reported in literature. Fe 3 O 4 is one of the cheaper magnetic oxide, which act in this study as a high surface area framework that is coated by the growth of Cu-apatite shell. The Fe 3 O 4 core has to facilitate the recycle of nanocatalysts through magnetic separation. This present work was achieved within our progressive program to develop an ecofriendly and efficient approach for the synthesis of various products. [46][47][48][49][50] This is the background upon which this work is situated upon, and the objective is based on the synthesis and characterization of Fe 3 O 4 @Cu-apatite nanoparticles, and then their use as magnetic catalysts for the N-arylation of imidazole with arylboronic acids (Scheme 1).

Materials and apparatus
Copper(II) nitrate (Cu(NO 3 ) 2 $3H 2 O, $98.0%), ferric chloride (FeCl 3 $6H 2 O), calcium nitrate (Ca(NO 3 ) 2 $4H 2 O, $99.0%), ammonium hydrogen phosphate ((NH 4 ) 2 HPO 4 ), sodium acetate (CH 3 COONa), ethylene glycol (EG), sodium hydroxide (NaOH), ammonia solution (NH 3 $H 2 O, 30%), poly (ethylene glycol) (PEG, 1000) were purchased from Aldrich Chemical Company and were used as received without any further purication. Deionized water was used in all experiments. Xray diffraction (XRD) patterns of the catalysts were obtained at room temperature on a Bruker AXS D-8 diffractometer using Cu-Ka radiation in Bragg-Brentano geometry (q-2q). SEM and STEM micrographs were obtained on a Tecnai G2 microscope at 120 kV. The average diameter of sample was quantied from the SEM image using ImageJ soware. The gas adsorption data was collected using a Micromeritics 3Flex Surface characterization analyzer, using nitrogen. Prior to nitrogen sorption, all samples were degassed at 150 C overnight. The specic surface areas were determined from the nitrogen adsorption/desorption isotherms (at À196 C), using the BET (Brunauer-Emmett-Teller) method. Pore size distributions were calculated from the N 2 adsorption isotherms with the "classic theory model" of Barret, Joyney and Halenda (BJH).
Fourier transformation infrared (FT-IR) spectra of samples in KBr pellets were measured on a Bruker Vector 22 spectrometer. Magnetic properties of Fe 3 O 4 and Fe 3 O 4 @Cu-apatite were investigated in a MPMS-XL-7AC superconducting quantum interference device (SQUID) magnetometer. The magnetic measurements were performed from À15 000 to 15 000 Oe at room temperature. The element content of Fe 3 O 4 @Cu-apatite materials was determined by an Elemental analysis was realized using inductively coupled plasma atomic emission spectrometry (ICP AES; Ultima 2 -Jobin Yvon). The 1 H NMR spectra were recorded in CDCl 3 or DMSO d6, using a Bruker Avance 600 spectrometer.

Catalyst preparation
2.2.1. Preparation of Fe 3 O 4 nanoparticles. The magnetic iron oxide nanoparticles used in this study were synthesized using a solvothermal method, adapting methodologies described in the literature. 51 In a typical procedure, 5.0 mmol of the FeCl 3 $6H 2 O was dissolved in 20 mL of ethylene glycol, and then 20 mmol of NaOH was added to the resultant mixture with vigorous stirring. Then, the obtained EG solution was transferred to a Teon-lined stainless-steel autoclave and sealed. Aer reacting at 180 C for 6 h, the autoclave was cooled down to ambient temperature and the mixture was withdrawn from the reactor with a magnet bar to get the Fe 3 O 4 magnetic NPs, which were then washed with water and ethanol and dried at 40 C for 24 h.
2.2.2. Fe 3 O 4 @Cu-apatite nanocomposite. The Fe 3 O 4 @Cuapatite nanocomposite was prepared along a hydrothermal route. 52 Firstly, 0.30 g of Fe 3 O 4 NPs was dispersed in 100 mL distilled water, and then 12 g of urea was added. Then, 10 mL of a solution produced from a mixture of Ca(NO 3 ) 2 $4H 2 O (0.5904 g) and Cu(NO 3 ) 2 $4H 2 O (0.604 g) was added into the above solution in drops (molar ratio of Ca : Cu 1 : 1). Aer heating at 90 C for 2 h, 10 mL of (NH 4 ) 2 HPO 4 (0.1981 g) solution was also added in drops and the pH value of the solution was tuned to 10 with ammonia water (molar ratio of Ca : P 10 : 6). Aer 0.5 h, the solution was transferred into a Teon-lined stainless-steel autoclave, where it was sealed and heated at 165 C for 12 h. Finally, the Fe 3 O 4 @Cu-apatite NPs were collected using a magnet bar.

General procedure of imidazol's N-arylation with arylboronic acid
In a 50 mL round-bottomed ask, imidazole (1 mmol), phenylboronic acid (1.2 mmol), K 2 CO 3 (1.5 mmol) and Fe 3 O 4 @Cuapatite (15 mol%) were added and stirred in MeOH under air at 60 C for the required time, monitoring by TLC. Aer completion, the mixture was diluted with H 2 O and the product was extracted with EtOAc (3 times). The combined extracts were washed with brine (3 times) and dried over Na 2 SO 4 . The product was puried using column chromatography (60-120 mesh silica gel, eluting with EtOAc-hexane). The structures of the prepared products were conrmed by 1 H NMR and assigned on the basis of their spectral data in comparison with those reported in the literature.

Catalyst characterization
XRD was used to analyze the crystalline phases of the asprepared Fe 3 O 4 and Fe 3 O 4 @Cu-apatite samples (Fig. 1). The      Table 1. The calculation of the BET surface area for Fe 3 O 4 is 22 m 2 g À1 and the pore volume is 0.0305 cm 3 g À1 . The surface area and pore volume of the Fe 3 O 4 @Cu-apatite are 84 m 2 g À1 and 0.1350 cm 3 g À1 , respectively. This change in surface properties was assumed to be due to the coating of Cu-apatite on the surface of Fe 3 O 4 nanoparticles. The rough surface provides more pores for nitrogen adsorption. The nitrogen sorption isotherm of Fe 3 O 4 is type IV, displaying a hysteresis loops type H3, indicating the mesoporous nature of the material (Fig. 3a). For the Fe 3 O 4 @Cuapatite, the nitrogen sorption isotherm is type (IV) with a hysteresis loop type H3, which proves the existence of mesopores according to the IUPAC manifests (Fig. 3b).
The corresponding pore size distribution curve indicates that Fe 3 O 4 and Fe 3 O 4 @Cu-apatite have a centralized pore size distribution around 4 and 11 nm, respectively, which corresponding to mesoporous materials (inset Fig. 3a and b).
SEM was used to study the surface morphology of Fe 3 O 4 and Fe 3 O 4 @Cu-apatite (Fig. 4) Fig. 5a and b, it can be observed that Fe 3 O 4 NPS show a very good dispersion of spherical particles. These Fe 3 O 4 spheres were further used as cores for the growth of Cu-apatite shells to obtain the Fe 3 O 4 @Cu-apatite core-shell nanostructures. Compared to the Fe 3 O 4 cores, the outside surfaces became coarse aer the growth process, as shown in Fig. 5c and d. These results indicate that the Cu-apatite nanoparticles were successfully loaded onto the Fe 3 O 4 sphere surfaces, which clearly demonstrates the formation of core-shell nanostructures. We have also noticed that Cu-apatite was still tightly anchored on the surface of Fe 3 O 4 even aer the severe conditions under which the sample preparation for STEM analysis were conducted (a long time of mechanical stirring and sonication), suggesting the existence of strong interactions between Cu-apatite and Fe 3 O 4 .
The magnetic properties of Fe 3 O 4 and Fe 3 O 4 @Cu-apatite were investigated using SQUID from À15 000 to 15 000 Oe at room temperature. As shown in Fig. 6    This superparamagnetism can make the magnetic nanoparticles dispersible in the solution with negligible magnetic interactions between each other, which avoids magnetic clustering. As shown in the Fig. 7, Fe 3 O 4 @Cu-apatite can be dispersed in deionized water to form a stable brown suspension before magnetic separation. However, when a magnet was placed close to the reaction vessel for a while, it could be observed that the synthesized samples were rapidly attracted to the magnet side, and a nearly colorless solution was obtained.

N-Arylation of imidazole over Fe 3 O 4 @Cu-apatite catalyst
Firstly, the activities of some samples were screened, such as: Fe 3 O 4 , Cu-apatite, Fe 3 O 4 @Cu-apatite and Cu-apatite@Fe 3 O 4 catalyzing imidazole N-arylation using phenyl boronic acid as a model reaction. The resultant obtained results are summarized in Table 2. First of all, the reaction doesn't occur without the addition of a catalyst ( Table 2, entry 1), and the iron oxide is also inactive in this reaction (Table 2, entry 2). However, the reaction of N-arylation of imidazole was occurring when Cuapatite was used as catalyst with a yield of 98% (Table 2, entry 3). These results can be explained by the presence of the copper, which play an important role as catalytic element for enhancing the cross-coupling reaction. Additionally, coating iron oxide by copper-apatite (Fe 3 O 4 @Cu-apatite) in order to separate them by the simple application of a magnet has no signicant effect on its catalytic activity ( Table 2, entry 4) with a yield of 97%. On the other hand, coating Cu-apatite by Fe 3 O 4 decrease their catalytic activity ( Table 2, entry 5), this can be explain by the decrease of the contact surface of the catalytic element and the reactants. This result shows the importance of the catalytic system developed in this work (Fe 3 O 4 @Cu-apatite) for the N-arylation of imidazole.
The inuence of various reaction parameters, such as type of solvent, nature of the base, reaction temperature, and catalyst    Table 3. Initially, with 20 mol% of the catalyst and K 2 CO 3 as a base, the screening of the effect of different solvents showed that H 2 O may not be the optimal solvent, when the reaction was carried out in EtOH or MeOH, the yield of the product 1-phenyl-1H-imidazole was isolated in 87 and 97%, respectively (Table 3, entry 1 and 3). Aer optimizing the solvent, the effect of the base was studied (Table 3, entries 4-6), and it was found that the best yields were obtained by using K 2 CO 3 as base. Otherwise, it was noted that the presence of the base was necessary for the achievement of the reaction as reported in the literature. This is explained by the role of the base to activate the boronic acid in the mechanism of the reaction. The study of the inuence of temperature, in the case of using MeOH, indicated that this reaction is very sensitive to changes in temperature (Table 3, entries 7-10). According to these studies we can conclude that temperature is a key parameter in this type of reaction, and the optimum yield was obtained at 60 C. Furthermore, the decreasing the catalyst loading to half results in a yield of 84% (Table 3, entry 12), which demonstrate the inuence of the catalyst amount in this type of reactions. Thus, the optimum conditions for this reaction are at 60 C in MeOH in the presence of 15 mol% of nanocatalyst with respect to imidazole and K 2 CO 3 (1 mmol) as the base (Table 3, entry 11).
With the optimized reaction conditions in hand, a variety of arylboronic acids were chosen as the substrates in this Nimidazole cross-coupling reaction and the results are shown in Table 4. The substitution effects of arylboronic acid were investigated. It was found that the reaction with phenylboronic acids with an electron donating group afforded better yields ( Table 4, entries 1-5) than with electron withdrawing groups ( Table 4, entries 6). Similar observation was made when indole was used in place of imidazoles to obtain the corresponding Narylindole (Table 4, entries [8][9][10]. Another benet of this nanocatalyst system consists of its ease of recyclability. The coupling of imidazole with arylboronic acid was chosen as a model reaction for the reusability study (Fig. 8). Aer the cross-coupling reaction, the recovered catalyst was washed twice by dichloromethane and water, dried at room temperature before being reused under similar conditions for the next run. It has to be noted that the catalytic behavior of the Fe 3 O 4 @Cu-apatite catalyst remains nearly the same for the ve successive runs. Aer the 5th cycle, a notable loss of the catalytic activity was observed. The N-arylation of N-arylheterocycles over Fe 3 O 4 @Cu-apatite was compared with other catalysts reported in the literature as tabulated in Table 5. From this, we can show clearly that our Fe 3 O 4 @Cu-apatite catalyst exhibited a best catalytic activity of N-arylation of imidazole comparing with other catalytic systems.
In order to investigate the behavior of the Fe 3 O 4 @Cu-apatite catalyst during recycling experiments, XRD and STEM analysis were performed (Fig. 9). It was observed from the XRD pattern of the Fe 3 O 4 @Cu-apatite catalyst that, aer one catalytic cycle, the crystalline composition of the catalyst remained unchanged.   Also, from STEM analysis, Cu-apatite was still tightly anchored on the surface of Fe 3 O 4 aer one catalytic cycle. The catalytic effect of copper ions leached from the Fe 3 O 4 @Cu-apatite was studied in the N-arylation reaction, which was carried out under the optimum conditions. Aer 2 hours of heating, the catalyst was removed and the remaining reaction mixture was reheated. It was observed that the concentration of the desired product didn't increase even aer heating for an additional 8 h. To further evaluate the catalytic effect of copper ions leached from the Fe 3 O 4 @Cu-apatite, ICP analysis was used against the catalyst before and aer reaction. The copper concentration of the catalyst was found to be 1.83 wt% for the fresh catalyst and 1.82 wt% aer one catalytic cycle in N-arylation of imidazole, which conrms negligible copper leaching.

Conclusion
In summary, magnetic Fe 3 O 4 @Cu-apatite core-shell nanocatalysts have been successfully prepared using a hydrothermal method. The physico-chemical properties of these materials were evaluated by FT-IR, XRD, SEM, STEM as well as the adsorptiondesorption of nitrogen. The prepared catalysts showed potential capability for the preparation of N-arylimidazoles through the Narylation reaction under mild conditions. Thereaer, the optimal reaction conditions were explored, and methanol was identied as the optimum solvent of the reaction, which is environmentally friendly. Furthermore, the Fe 3 O 4 @Cu-apatite can also be easily separated by an external magnet and reused for ve runs with only a slight decrease in its catalytic activity.

Conflicts of interest
The authors declare that they have no competing interests.