Novel Magnetically-Recyclable, Nitrogen-Doped Fe3O4@Pd NPs for Suzuki–Miyaura Coupling and Their Application in the Synthesis of Crizotinib

1 School of Chemistry and Environmental Engineering, Jiangsu University of Technology, Changzhou 213001, China; zkai86@163.com 2 College of Biology and Environmental Engineering, Zhejiang Shuren University, Hangzhou 310015, China; tjy132@msn.com (J.T.); hzjjz@163.com (J.J.) 3 College of Material Chemistry and Chemical Engineering, Hangzhou Normal University, Hangzhou, 310036, China; pfzhang@hznu.edu.cn * Correspondence: shenchaozju@163.com (C.S.); workhard84@126.com (J.Q.); Tel.: +86‐0571‐88297172 (C.S.); +86‐0571‐88297103 (J.Q.)


Introduction
Catalysts are gaining increasing importance, due to their effective manner of solving energy and resource problems, which have become an important part of achieving sustainable development strategies in the 21st century [1,2].Among the noble metals, palladium and nickel are the most useful catalysts for the formation of C-C bonds in organic transformations [3][4][5].In the past, homogeneous Pd catalysts made significant progress in Suzuki-Miyaura coupling reactions; however, it is difficult to separate the products and reuse them.To overcome the drawbacks of homogeneous catalysts, heterogeneous catalysts were significantly explored [6][7][8][9][10].Heterogeneous catalysis has some advantages such as a recyclable catalytic systems, nontoxic ligands, and a lower amount of palladium residues in products [11][12][13][14].The recovery of Pd catalysts from reaction systems is not easy, and attempts to solve the problem were made by immobilizing the active metal species on supports, such as carbon, silica, metal oxide, polymer, and nanocomposites [15][16][17][18][19][20].The magnetic core/shell-supported catalyst is an excellent solution, and its intrinsic magnetic properties enable the efficient separation of the catalysts from the reaction system with an external magnetic field [21,22].For example, Kumar et al. reported the use of Fe 3 O 4 @C/Pd as an excellent catalyst for the hydrogenation of aromatic nitro compounds, Suzuki coupling, and sequential reactions; the reactions worked well and gave excellent yields [23].Sun and coworkers found that the magnetic Fe 3 O 4 @C/Pd catalyst showed high catalytic activity for the yield of biphenyl, and it could maintain 90% activity even after being recycled ten times [24].Fang et al. demonstrated that magnetic Fe 3 O 4 @C/Pd microsphere catalysts were active in Suzuki coupling [25].
Nitrogen-doped carbon has attracted much attention due to its special structure, good properties, and potential applications [26][27][28][29][30]. Zhang et al. reported that Pd@C-N's high catalytic performance is attributed to the unique structure of the catalytic support-metal and support-substrate junctions.Wang's group found that an as-synthesized Pd/N-carbon nanotube (CNT) catalyst showed high catalytic activity in the Heck reaction, and that the catalyst could be reused at least five times in the aerobic oxidation of benzyl alcohol [31,32].Movahed and coworkers demonstrated that the good reactivity of the Pd NP-high nitrogen-doped graphene (HNG) catalyst in Suzuki coupling was attributed to the high degree of nitrogen loading in graphene sheets [33].
Continuing our longstanding interest in developing novel carbohydrate-derived catalysts for C-C or C-S coupling reactions [34][35][36][37][38], we were interested in developing a green and efficient chemistry protocol for C-C coupling reactions and related practical applications.Herein, we describe the efficient synthesis of a magnetically-recyclable, nitrogen-doped Fe 3 O 4 @Pd catalyst for the Suzuki coupling of aryl or heteroaryl halides (I, Br, Cl) with arylboronic acids.These as-prepared catalysts could be easily isolated from the reaction mixture using an external magnet, and they could be reused at least ten times with excellent yields achieved.In addition, the marketed drug crizotinib (anti-tumor) could be easily synthesized using this protocol.

Results and Discussion
The preparation procedures of Fe 3 O 4 @C/Pd and Fe 3 O 4 @NC/Pd involve three steps, as shown in Scheme 1.Initially, the Fe 3 O 4 particles were prepared via a robust solvothermal reaction based on the high-temperature reduction of FeCl 3 •6H 2 O in ethylene glycol [39].Then, a thin carbon layer was modified with ethylenediamine (EDA) by stirring a mixture of Fe 3 O 4 , glucose, and EDA in water.The mixture was coated on the surface of the magnetite Fe 3 O 4 particles via carbonization under hydrothermal conditions [40].Ultimately, the Fe 3 O 4 @NC/Pd and Fe 3 O 4 @C/Pd catalysts were obtained upon adding PdCl 2 to Fe 3 O 4 @NC or Fe 3 O 4 @C in ethanol, followed by ascorbic-acid reduction, to generate Pd(0) nanoparticles.Fe3O4@C/Pd catalyst showed high catalytic activity for the yield of biphenyl, and it could maintain 90% activity even after being recycled ten times [24].Fang et al. demonstrated that magnetic Fe3O4@C/Pd microsphere catalysts were active in Suzuki coupling [25].Nitrogen-doped carbon has attracted much attention due to its special structure, good properties, and potential applications [26][27][28][29][30]. Zhang et al. reported that Pd@C-N's high catalytic performance is attributed to the unique structure of the catalytic support-metal and supportsubstrate junctions.Wang's group found that an as-synthesized Pd/N-carbon nanotube (CNT) catalyst showed high catalytic activity in the Heck reaction, and that the catalyst could be reused at least five times in the aerobic oxidation of benzyl alcohol [31,32].Movahed and coworkers demonstrated that the good reactivity of the Pd NP-high nitrogen-doped graphene (HNG) catalyst in Suzuki coupling was attributed to the high degree of nitrogen loading in graphene sheets [33].
Continuing our longstanding interest in developing novel carbohydrate-derived catalysts for C-C or C-S coupling reactions [34][35][36][37][38], we were interested in developing a green and efficient chemistry protocol for C-C coupling reactions and related practical applications.Herein, we describe the efficient synthesis of a magnetically-recyclable, nitrogen-doped Fe3O4@Pd catalyst for the Suzuki coupling of aryl or heteroaryl halides (I, Br, Cl) with arylboronic acids.These as-prepared catalysts could be easily isolated from the reaction mixture using an external magnet, and they could be reused at least ten times with excellent yields achieved.In addition, the marketed drug crizotinib (anti-tumor) could be easily synthesized using this protocol.

Results and Discussion
The preparation procedures of Fe3O4@C/Pd and Fe3O4@NC/Pd involve three steps, as shown in Scheme 1.Initially, the Fe3O4 particles were prepared via a robust solvothermal reaction based on the high-temperature reduction of FeCl3•6H2O in ethylene glycol [39].Then, a thin carbon layer was modified with ethylenediamine (EDA) by stirring a mixture of Fe3O4, glucose, and EDA in water.The mixture was coated on the surface of the magnetite Fe3O4 particles via carbonization under hydrothermal conditions [40].Ultimately, the Fe3O4@NC/Pd and Fe3O4@C/Pd catalysts were obtained upon adding PdCl2 to Fe3O4@NC or Fe3O4@C in ethanol, followed by ascorbic-acid reduction, to generate Pd(0) nanoparticles.Scheme 1. Synthesis of Fe3O4@C/Pd and Fe3O4@NC/Pd catalysts.These as-prepared catalysts were characterized by infrared analysis (FT-IR), thermogravimetric analysis (TG), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and inductively coupled plasma-optical emission spectrometry (ICP-OES).In Figure 1, the presence of absorption bands at 673 cm −1 could correspond to the absorption band of Fe-O and 1336 cm −1 could correspond to the O-H bending vibrations; 1558 cm −1 and 3386 cm −1 are associated with the C=O and O-H vibrations, which confirms the successful attachment of nitrogen-doped carbon on the surface of Fe3O4.This also reflects the carbonization of aminated glucose during the hydrothermal process and suggests the presence of large amounts of hydrophilic groups on the Fe3O4@NC [41].The FTIR spectra of Fe3O4@NC/Pd (c) and Fe3O4@NC (b) were similar to Fe3O4 (a), however the FTIR spectra of Fe3O4@C/Pd and Fe3O4@NC/Pd were different (Figure S1).The existence of absorption bands at 1382 cm −1 (C-H) and 1250 cm −1 (C-O) confirmed the arylide was adsorbed by Fe3O4@NC/Pd catalyst after the catalytic reaction (Figure S2).These as-prepared catalysts were characterized by infrared analysis (FT-IR), thermogravimetric analysis (TG), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD) and inductively coupled plasma-optical emission spectrometry (ICP-OES).In Figure 1, the presence of absorption bands at 673 cm −1 could correspond to the absorption band of Fe-O and 1336 cm −1 could correspond to the O-H bending vibrations; 1558 cm −1 and 3386 cm −1 are associated with the C=O and O-H vibrations, which confirms the successful attachment of nitrogen-doped carbon on the surface of Fe 3 O 4 .This also reflects the carbonization of aminated glucose during the hydrothermal process and suggests the presence of large amounts of hydrophilic groups on the Fe 3 O 4 @NC [41].The FTIR spectra of Fe 3 O 4 @NC/Pd (c) and Fe 3 O 4 @NC (b) were similar to Fe 3 O 4 (a), however the FTIR spectra of Fe 3 O 4 @C/Pd and Fe 3 O 4 @NC/Pd were different (Figure S1).
The existence of absorption bands at 1382 cm −1 (C-H) and 1250 cm −1 (C-O) confirmed the arylide was adsorbed by Fe 3 O 4 @NC/Pd catalyst after the catalytic reaction (Figure S2).The thermal stability of Fe3O4 (a), Fe3O4@NC (b) and Fe3O4@NC/Pd (c) was then proved by TG analysis.Figure 2 shows that catalysts were stable up to 250 °C and suggests that their high thermal stability allows them to be compatible with most organic reactions.The TEM image of Fe3O4@NC/Pd is shown in Figure 3.These microspheres essentially have a typical core/shell nanostructure and the average diameter of the catalyst was about 300 nm.The final morphology of Fe3O4@NC and Fe3O4@C shows a core-shell feature with Pd uniformly deposited on the surface.The TEM studies confirmed the incorporated Pd NPs on the surface of Fe3O4@NC nanospheres and also indicated that the catalyst had a core-shell nanostructure.In addition, the results mean the carbonization did not damage the core-shell structure.The thermal stability of Fe 3 O 4 (a), Fe 3 O 4 @NC (b) and Fe 3 O 4 @NC/Pd (c) was then proved by TG analysis.Figure 2 shows that catalysts were stable up to 250 • C and suggests that their high thermal stability allows them to be compatible with most organic reactions.The thermal stability of Fe3O4 (a), Fe3O4@NC (b) and Fe3O4@NC/Pd (c) was then proved by TG analysis.Figure 2 shows that catalysts were stable up to 250 °C and suggests that their high thermal stability allows them to be compatible with most organic reactions.The TEM image of Fe3O4@NC/Pd is shown in Figure 3.These microspheres essentially have a typical core/shell nanostructure and the average diameter of the catalyst was about 300 nm.The final morphology of Fe3O4@NC and Fe3O4@C shows a core-shell feature with Pd uniformly deposited on the surface.The TEM studies confirmed the incorporated Pd NPs on the surface of Fe3O4@NC nanospheres and also indicated that the catalyst had a core-shell nanostructure.In addition, the results mean the carbonization did not damage the core-shell structure.The TEM image of Fe 3 O 4 @NC/Pd is shown in Figure 3.These microspheres essentially have a typical core/shell nanostructure and the average diameter of the catalyst was about 300 nm.The final morphology of Fe 3 O 4 @NC and Fe 3 O 4 @C shows a core-shell feature with Pd uniformly deposited on the surface.The TEM studies confirmed the incorporated Pd NPs on the surface of Fe 3 O 4 @NC nanospheres and also indicated that the catalyst had a core-shell nanostructure.In addition, the results mean the carbonization did not damage the core-shell structure.The thermal stability of Fe3O4 (a), Fe3O4@NC (b) and Fe3O4@NC/Pd (c) was then proved by TG analysis.Figure 2 shows that catalysts were stable up to 250 °C and suggests that their high thermal stability allows them to be compatible with most organic reactions.The TEM image of Fe3O4@NC/Pd is shown in Figure 3.These microspheres essentially have a typical core/shell nanostructure and the average diameter of the catalyst was about 300 nm.The final morphology of Fe3O4@NC and Fe3O4@C shows a core-shell feature with Pd uniformly deposited on the surface.The TEM studies confirmed the incorporated Pd NPs on the surface of Fe3O4@NC nanospheres and also indicated that the catalyst had a core-shell nanostructure.In addition, the results mean the carbonization did not damage the core-shell structure.As shown in Figure 4, the electronic properties of the Fe 3 O 4 @NC/Pd catalyst were probed by XPS analysis.As shown in Figure 4a, the peaks corresponding to C1s, N 1s, O 1s, and Pd 3s, 3p and 3d were clearly observed in the XPS survey spectroscopy.This indicates that the nitrogen successfully doped the Fe 3 O 4 @Pd NPs.The C1s peak of Fe 3 O 4 @NC/Pd catalyst is shown in Figure 4b.The main peak at 281.1 eV was associated with the C-C, implying that most of the carbon atoms in the Fe 3 O 4 @NC/Pd catalyst were arranged in a conjugated honeycomb lattice.As shown in Figure 4c, the binding energy of Pd 3d 3/2 and Pd 3d 5/2 were 331.1 eV and 336.2 eV, respectively.The two peaks were the characteristic peaks of Pd(0), suggesting that the absorbed Pd(II) was successfully reduced to Pd(0) nanoparticles under ascorbic-acid reduction.By ICP-OES detection, the content of the Pd element loaded on Fe 3 O 4 @NC/Pd catalyst was found to be 5 wt%.As shown in Figure 4, the electronic properties of the Fe3O4@NC/Pd catalyst were probed by XPS analysis.As shown in Figure 4a, the peaks corresponding to C1s, N 1s, O 1s, and Pd 3s, 3p and 3d were clearly observed in the XPS survey spectroscopy.This indicates that the nitrogen successfully doped the Fe3O4@Pd NPs.The C1s peak of Fe3O4@NC/Pd catalyst is shown in Figure 4b.The main peak at 281.1 eV was associated with the C-C, implying that most of the carbon atoms in the Fe3O4@NC/Pd catalyst were arranged in a conjugated honeycomb lattice.As shown in Figure 4c, the binding energy of Pd 3d3/2 and Pd 3d5/2 were 331.1 eV and 336.2 eV, respectively.The two peaks were the characteristic peaks of Pd(0), suggesting that the absorbed Pd(II) was successfully reduced to Pd(0) nanoparticles under ascorbic-acid reduction.By ICP-OES detection, the content of the Pd element loaded on Fe3O4@NC/Pd catalyst was found to be 5 wt%.To evaluate the catalytic performance of these catalysts, Suzuki coupling were carried out as model reactions.The reactions were carried out using water as the solvent and by coupling 4-iodoanisole (1a) with phenylboronic acid (2a), and using different reaction parameters such as the base, the temperature, the time, the dosage and the kind of catalyst to obtain the best reaction conditions.As shown in Table 1, the Fe3O4 and the magnetic core-carbon shell were not able to catalyze the reaction (Table 1, entries 1-3).Fe3O4@C/Pd and Fe3O4@NC/Pd showed good results due to their high reactivity, demonstrating, respectively, that N-doped carbon has a great influence on  To evaluate the catalytic performance of these catalysts, Suzuki coupling were carried out as model reactions.The reactions were carried out using water as the solvent and by coupling 4-iodoanisole (1a) with phenylboronic acid (2a), and using different reaction parameters such as the base, the temperature, the time, the dosage and the kind of catalyst to obtain the best reaction conditions.As shown in Table 1, the Fe 3 O 4 and the magnetic core-carbon shell were not able to catalyze the reaction (Table 1, entries 1-3).Fe 3 O 4 @C/Pd and Fe 3 O 4 @NC/Pd showed good results due to their high reactivity, demonstrating, respectively, that N-doped carbon has a great influence on the catalytic results, and an increase in the nitrogen loading in carbon sheet leads to an increase in the product yield (Table 1, entries 4, 5).Then, various bases such as K 2 CO 3 , NaOH, Na 2 CO 3 , KOH, Et 3 N and Cs 2 CO 3 were also screened for their effect on the reaction (Table 1, entries 5-10); the best yield was obtained when KOH was used (Table 1, entry 8).Besides, the effect of different temperatures was explored and the results showed that 90 • C was more appropriate for the Suzuki coupling than other temperatures and provided the highest yield of 96% (Table 1, entries 8, 11-14).Next, the effect of the catalyst dosage was examined and the best result was obtained when 10 mg Pd was used as the catalyst (Table 1, entries 8, 16-18).Finally, it was found that the reaction after 0.5 h resulted in a higher yield (Table 1, entries 8, 19-20).the catalytic results, and an increase in the nitrogen loading in carbon sheet leads to an increase in the product yield (Table 1, entries 4, 5).Then, various bases such as K2CO3, NaOH, Na2CO3, KOH, Et3N and Cs2CO3 were also screened for their effect on the reaction (Table 1, entries 5-10); the best yield was obtained when KOH was used (Table 1, entry 8).Besides, the effect of different temperatures was explored and the results showed that 90 °C was more appropriate for the Suzuki coupling than other temperatures and provided the highest yield of 96% (Table 1, entries 8, 11-14).
Next, the effect of the catalyst dosage was examined and the best result was obtained when 10 mg Pd was used as the catalyst (Table 1, entries 8, 16-18).Finally, it was found that the reaction after 0.5 h resulted in a higher yield (Table 1, entries 8, 19-20).After obtaining the optimal reaction conditions, the substrate scope of the aryl halides and arylboronic acids was studied.As shown in Table 2, the effect of different aryl iodides was first carried out using phenylboronic acid (2a) as a substrate.The result showed that the substrates with electron-releasing groups gave good yields when compared to electron-withdrawing groups in aryl halides (Table 2, entries 1-13) and meta-substituted or ortho-substituted substrates showed lower yields than the para-substituted arylhalides (Table 2, entries 2,4-7,10-13).These results showed that the steric hindrance and electronic effect of substrates 1a-m had little effect on the Suzuki coupling under the optimized reaction conditions.Then, the efficiency of the protocol for the Suzuki coupling of aryl bromides or chlorides with corresponding boronic acids was examined.The reaction conditions were quite effective for the coupling of aryl bromides with boronic acids, resulting in high yields (Table 2, entries 14-16).However, only a moderate yield were obtained when aryl chlorides were used as the substrate (Table 2, entries [17][18][19][20].The coupling reactions using arylboronic acids were also investigated, and the coupling products were obtained in good yields, After obtaining the optimal reaction conditions, the substrate scope of the aryl halides and arylboronic acids was studied.As shown in Table 2, the effect of different aryl iodides was first carried out using phenylboronic acid (2a) as a substrate.The result showed that the substrates with electron-releasing groups gave good yields when compared to electron-withdrawing groups in aryl halides (Table 2, entries 1-13) and meta-substituted or ortho-substituted substrates showed lower yields than the para-substituted arylhalides (Table 2, entries 2,4-7,10-13).These results showed that the steric hindrance and electronic effect of substrates 1a-m had little effect on the Suzuki coupling under the optimized reaction conditions.Then, the efficiency of the protocol for the Suzuki coupling of aryl bromides or chlorides with corresponding boronic acids was examined.The reaction conditions were quite effective for the coupling of aryl bromides with boronic acids, resulting in high yields (Table 2, entries 14-16).However, only a moderate yield were obtained when aryl chlorides were used as the substrate (Table 2, entries [17][18][19][20].The coupling reactions using arylboronic acids were also investigated, and the coupling products were obtained in good yields, the electron-releasing groups in arylboronic acid gave higher yields compared to substrates bearing electron-withdrawing groups (Table 2, entries [21][22][23][24][25][26].Moreover, the optimized reaction conditions were effective for the Suzuki coupling of heteroaryl bromides with phenylboronic acids and produced products in satisfactory yields (Table 2, entries 27-32).the electron-releasing groups in arylboronic acid gave higher yields compared to substrates bearing electron-withdrawing groups (Table 2, entries 21-26).Moreover, the optimized reaction conditions were effective for the Suzuki coupling of heteroaryl bromides with phenylboronic acids and produced products in satisfactory yields (Table 2, entries 27-32).a The reaction conditions: aryl halides 1 (1 mmol), arylboronic acid 2 (1.5 mmol), Fe3O4@NC/Pd catalyst (10 mg), and KOH (1.5 mmol) in water (3 mL) under air.b Isolated yield.c 8 h.
With this methodology in hand, we turned our attention to the preparation of crizotinib, which is a potent and selective Me-senchymal epithelial factor/anaplastic lymphoma kinase (c-Met/ALK) inhibitor (Scheme 2) [42].Crizotinib has a palladium residue problem because the aminopyridine coordinates to palladium to form the corresponding stable compounds.As a result, the separation of the crizotinib API from the residual palladium has been a challenging task [43].Thus, the coupling between aryl bromide 4 and pinacol boronate 5 were carried out employing this method.This  With this methodology in hand, we turned our attention to the preparation of crizotinib, which is a potent and selective Me-senchymal epithelial factor/anaplastic lymphoma kinase (c-Met/ALK) inhibitor (Scheme 2) [42].Crizotinib has a palladium residue problem because the aminopyridine coordinates to palladium to form the corresponding stable compounds.As a result, the separation of the crizotinib API from the residual palladium has been a challenging task [43].Thus, the coupling between aryl bromide 4 and pinacol boronate 5 were carried out employing this method.This transformation was accomplished with excellent results in the presence of the Fe 3 O 4 @NC/Pd catalyst and KOH in water at 90 • C for 6 h.Then the intermediate 6 was treated with 4 M HCl in 1,4-dioxane/CH 2 Cl 2 , and the crizotinib API was isolated in very high yield (95%) with >99% purity and <10 ppm Pd.
transformation was accomplished with excellent results in the presence of the Fe3O4@NC/Pd catalyst and KOH in water at 90 °C for 6 h.Then the intermediate 6 was treated with 4 M HCl in 1,4-dioxane/CH2Cl2, and the crizotinib API was isolated in very high yield (95%) with >99% purity and <10 ppm Pd.The recyclability of the catalyst was then studied using the Suzuki coupling reaction.The reuse experiments for the Fe3O4@NC/Pd catalyst were carried out and the catalyst was able to be separated by a permanent magnet after each round and reused in next catalytic reaction.As shown in Figure 5, the magnetic catalyst remained effective and stable after the tenth round, affording a coupling product with 90% yield, which indicated the good stability of the Fe3O4@NC/Pd catalyst.It is well known that using magnetic separation to recycle catalysts is much easier than filtration and centrifugation.A key factor to be investigated is the stability of the catalyst: the leaching of active species into the reaction mixture.When exploring the leaching of Pd from the catalyst, the concentration of Pd in the Fe3O4@NC/Pd catalyst was found to be unchanged using ICP-OES analysis.A hot filleting leaching experiment was also conducted.After 0.5 h, the magnetic The recyclability of the catalyst was then studied using the Suzuki coupling reaction.The reuse experiments for the Fe 3 O 4 @NC/Pd catalyst were carried out and the catalyst was able to be separated by a permanent magnet after each round and reused in next catalytic reaction.As shown in Figure 5, the magnetic catalyst remained effective and stable after the tenth round, affording a coupling product with 90% yield, which indicated the good stability of the Fe 3 O 4 @NC/Pd catalyst.
transformation was accomplished with excellent results in the presence of the Fe3O4@NC/Pd catalyst and KOH in water at 90 °C for 6 h.Then the intermediate 6 was treated with 4 M HCl in 1,4-dioxane/CH2Cl2, and the crizotinib API was isolated in very high yield (95%) with >99% purity and <10 ppm Pd.The recyclability of the catalyst was then studied using the Suzuki coupling reaction.The reuse experiments for the Fe3O4@NC/Pd catalyst were carried out and the catalyst was able to be separated by a permanent magnet after each round and reused in next catalytic reaction.As shown in Figure 5, the magnetic catalyst remained effective and stable after the tenth round, affording a coupling product with 90% yield, which indicated the good stability of the Fe3O4@NC/Pd catalyst.It is well known that using magnetic separation to recycle catalysts is much easier than filtration and centrifugation.A key factor to be investigated is the stability of the catalyst: the leaching of active species into the reaction mixture.When exploring the leaching of Pd from the catalyst, the concentration of Pd in the Fe3O4@NC/Pd catalyst was found to be unchanged using ICP-OES analysis.A hot filleting leaching experiment was also conducted.After 0.5 h, the magnetic It is well known that using magnetic separation to recycle catalysts is much easier than filtration and centrifugation.A key factor to be investigated is the stability of the catalyst: the leaching of active species into the reaction mixture.When exploring the leaching of Pd from the catalyst, the concentration of Pd in the Fe 3 O 4 @NC/Pd catalyst was found to be unchanged using ICP-OES analysis.A hot filleting leaching experiment was also conducted.After 0.5 h, the magnetic Fe 3 O 4 @NC/Pd catalyst was collected by an external magnet, and the clear liquid solution was continuously stirred at 90 • C for 3 h and 2 ppm palladium was determined in the reaction solution, which indicated that there was no leaching of palladium from the catalyst to the reaction solution.

Experimental Materials
The starting materials were commercially available and were used without further purification except for the solvents.Ferric chloride hexahydrate (FeCl 3 •6H 2 O) was provided by Shanghai Darui Fine Chemicals Co. Ltd (Shanghai, China).Glucose was obtained from Chinasun Specialty Products Co. Ltd (Changshu, China).Sodium acetate anhydrous and ethylene glycol were provided by Shanghai Ling Feng Chemical Reagent Co. Ltd (Shanghai, China).Polyvinyl pyrrolidone (PVP) and ethylenediamine (EDA) were supplied by Aladdin (Shanghai, China).Palladium(II) chloride (PdCl 2 , 59.5%) was provided by J&K Scientific Ltd (Shanghai, China).Other materials were of analytical grade and used as received.

Characterization
Fourier-transform infrared (FTIR) spectra were determined with a Bruker Tensor 27 FT-IR (Swiss Brüker, Hangzhou, China) using KBr pellets.Melting points were measured on an X-6 Data microscopic melting point apparatus.Transmission electron microscopy (TEM) images were obtained from a FEI T20 microscope (FEI, Shanghai, China).X-ray diffraction (XRD) measurements were obtained using a Shimadzu XRD-6000 spectrometer (Shimadzu Corp, Beijing, China).X-ray photoelectron spectrographs (XPS) were determined with an Axis Ultra DLD electron spectrometer (Kratos Analytical, Manchester, UK) with the C1s = 284.8eV signal as the internal standard.The magnetic properties of the catalysts were measured using a vibrating sample magnetometer (VSM) (Lake Shore, New York, NY, USA).Thermogravimetric analyses (TG) were performed with a Q600 simultaneous DSC-TGA (TA Instruments, Shanghai, China) at 20 • C/min in a nitrogen atmosphere (150 mL/min).A total of 10 mg of each sample in an alumina pan was analyzed in the 30-800 • C temperature range. 1 H and 13 C NMR spectra were measured with a Bruker Advance 400 spectrometer (Swiss Brüker, Hangzhou, China).by using CDCl 3 or DMSO-d 6 as solvents and TMS as the internal standard.The Pd content in the catalyst was measured using a Perkin-Elmer Optima 2100 DV (PerkinElmer, Shanghai, China).

Preparation of Fe 3 O 4 Nanoparticles
Fe 3 O 4 nanoparticles were prepared using a solvothermal reaction method [44].Typically, FeCl 3 •6H 2 O (1.5 g), NaAC (2 g) and PVP (1 g) were dissolved in ethylene glycol (30 mL) under magnetic stirring.The resultant solution was transferred into a Teflon-lined stainless steel autoclave, sealed, and heated to 200 • C for 12 h.After completion of the reaction, the resulting product was separated using an external magnet and washed several times with ethanol water.Finally, the black Fe 3 O 4 nanoparticles were dried under vacuum at 60 • C for 24 h.

Preparation of Fe 3 O 4 @C Nanoparticles
Fe 3 O 4 @C was prepared by the in situ carbonization of glucose in the presence of Fe 3 O 4 under hydrothermal conditions [45].Fe 3 O 4 nanoparticles (200 mg) were dispersed in water (10 mL) containing glucose (3.2 g) by ultrasonication.Subsequently, it was put into a Teflon lined stainless steel autoclave, sealed, and heated at 180 • C for 10 h and cooled down at room temperature.After completion of the reaction, the resulting nanoparticles were obtained by an external magnet and washed with ethanol followed by water.Finally, the black colored product was dried under vacuum for 24 h to produce Fe 3 O 4 @C nanoparticles.

Preparation of Fe 3 O 4 @NC Nanoparticles
Fe 3 O 4 @NC was synthesized according to the procedure described in the literature [46].Fe 3 O 4 nanoparticles (100 mg) were first put into 10 mL water containing ethylenediamine (EDA) (0.2 mL) and glucose (1.6 g) and ultrasonicated.Subsequently, the mixture was placed into a Teflon-lined stainless steel autoclave, then the autoclave was sealed and heated at 180 • C for 10 h, before cooling to room temperature.After completion of the reaction, the product was obtained by an external magnet and

Scheme 2 .
Scheme 2. Application of the catalyst in the synthesis of crizotinib.

Figure 5 .
Figure 5. Recycling and reuse of Fe3O4@NC/Pd in the Suzuki coupling.

Scheme 2 .
Scheme 2. Application of the catalyst in the synthesis of crizotinib.

Scheme 2 .
Scheme 2. Application of the catalyst in the synthesis of crizotinib.

Figure 5 .
Figure 5. Recycling and reuse of Fe3O4@NC/Pd in the Suzuki coupling.

Figure 5 .
Figure 5. Recycling and reuse of Fe 3 O 4 @NC/Pd in the Suzuki coupling.

Table 1 .
Optimization of reaction conditions.a.

Table 1 .
Optimization of reaction conditions.a.

Table 2 .
Suzuki coupling between aryl halides and arylboronic acids in the presence of Fe 3 O 4 @NC/Pd .

Table 2 .
Suzuki coupling between aryl halides and arylboronic acids in the presence of Fe3O4@NC/Pd.a.