2.1 Chemicals

All solvents and chemicals (analytical grades) were purchased and used without further purification. Palladium(II) chloride (99.9%) was purchased from the Merck Company. Fourier transform infrared (FT-IR) spectra were obtained in an area of 400–4000 cm⁻¹ using a Nicolet IR100 instrument with spectroscopic grade KBr. The images and EDX analyses of the catalyst were observed using gold-coated Philips XL 30 and S-4160 instruments equipped with dispersive X-ray spectroscopy capabilities. Thermogravimetric analysis was performed on a thermal analyzer with a heating rate of 10 °C min⁻¹ over a temperature region of 25–600 °C under flowing compressed nitrogen gas. Powder X-ray diffraction (XRD) spectra were recorded at room temperature using a Philips X-Pert 1710 diffractometer and Co Kα (λ= 1.78897 Å) at a voltage of 40 kV and current of 40 mA. Data were collected from 10° to 90° (2θ) with a scan speed of 0.02° s⁻¹. The magnetic properties of the catalyst nanoparticles were obtained using a vibrating magnetometer/alternating gradient force magnetometer. The morphology of the catalyst was studied using scanning electron microscopy (SEM; Zeiss SIGMA VP) and transmission electron microscopy (TEM; Zeiss-EM10C-100 kV).

2.2 Preparation of sepiolite-coated magnetic nanoparticles (Fe₃O₄@sepiolite)

To prepare the sepiolite magnetic nanoparticles (Fe₃O₄@sepiolite), a mixture of 1 g sepiolite was dispersed in 50 mL deionized water and stirred for 4 h in an ultrasonic bath. This suspension was added to a mixture of 0.994 g (5 mmol) FeCl₂·4H₂O and 2.7 g (10 mmol) FeCl₃·6H₂O salts, which were dissolved in 100 mL of deionized water under intense stirring. An aqueous ammonia solution (30 mL) (25% w/w) was added to the resultant mixture to increase the reaction pH to about 11. The mixture was stirred at 80 °C for 1 h. After cooling to room temperature, the sepiolite-coated magnetic nanoparticle composites were collected with an external magnet, washed three times in hot water, and dried at 100 °C for 12 h.

2.3 Synthesis of palladium ion encapsulated with sepiolite-coated magnetic nanoparticles (Fe₃O₄@sepiolite-Pd²⁺)

Palladium chloride 177 mg (1 mmol) was added to 1 g Fe₃O₄@sepiolite in 50 mL deionized water, and the mixture was stirred at room temperature for 24 h. Subsequently, the final composite was separated with an external magnet, washed three times with deionized water, and dried at 80 °C for 12 h (Scheme 1).

2.4 General procedure for reduction of nitroarenes in the presence of Fe₃O₄@sepiolite-Pd²⁺

To the stirred mixture of nitroarene compound (1 mmol) in 4 mL EtOH/H₂O (3:1), 5 mmol NaBH₄ (0.189 g) and 20 mg of Fe₃O₄@sepiolite-Pd²⁺ were added. The reaction vessel was heated at 70 °C in an oil bath for 10–50 min. The progress of the reaction was scanned by thin-layer chromatography (TLC). After completion of the reaction, the nanoparticles were separated from the reaction mixture by an external magnet and washed repeatedly with deionized water and diethyl ether. Water (50 mL) was then added to the reaction mixture, extracted with CH₂Cl₂ (2 × 20 mL), and dried over Na₂SO₄. The solvent was removed under reduced pressure, and the products were purified with silica-gel plate chromatography using n-hexane–ethyl acetate as an eluent to afford pure product.

2.5 General procedure for the Suzuki coupling reaction in the presence of Fe₃O₄@sepiolite-Pd²⁺

Aryl halide compound (1 mmol) and K₂CO₃ (2 mmol) in EtOH/H₂O (1:1, 8 mL) and 10 mg of Fe₃O₄@sepiolite-Pd²⁺ were added to a stirred mixture of phenylboronic acid (1 mmol). The mixture was then heated and stirred at 70 °C for 15 min. The progress of the reaction was monitored by TLC. After completion of the reaction, the nanoparticles were separated from the reaction mixture with an external magnet and washed repeatedly with deionized water and diethyl ether. Water (50 mL) was then added to the reaction mixture, extracted with CH₂Cl₂ (2 × 20 mL), and dried over Na₂SO₄. The solvent was removed under reduced pressure, and the products were purified using a column chromatography eluted with n-hexane. The solvent was removed to give pure product.

3.1 Characterization of Fe₃O₄@sepiolite-Pd²⁺

The FT-IR spectrum of sepiolite (a), Fe₃O₄@sepiolite (b), and Fe₃O₄@sepiolite-Pd²⁺ (c) is depicted in Fig. 2. The spectrum of pristine sepiolite is shown in Fig. 2(a). The absorption bands at 3686 and 3562 cm⁻¹ correspond to the OH groups of Mg₃OH. The stretching vibration at 1658 cm⁻¹ can be attributed to the water of zeolite, and the bands at the 1015 and 470 cm⁻¹ can be assigned as the Si–O–Si vibration [16]. Fig. 2(b) shows the FT-IR spectrum of sepiolite-supported magnetic nanoparticles (Fe₃O₄@sepiolite). The typical bands that appear at 400–600 cm⁻¹ are caused by the FeO stretching vibration. The absorption band at 3562 cm⁻¹ in the IR spectrum of sepiolite is replaced by a broadband at 3441 cm⁻¹, which is because of the stretching vibrations of the hydroxyl groups in Fe₃O₄ and water. This shift proves the binding of a sepiolite to the surface of Fe₃O₄ nanoparticles through the chemisorption of the hydroxyl group [17].

The FT-IR spectrum of the final composite (Fe₃O₄@sepiolite-Pd²⁺), which was prepared from the stabilization of PdCl₂ on Fe₃O₄@sepiolite, is indicated in Fig. 2(c). The hydroxyl group stretching frequency is shifted to a lower frequency at 3418 cm⁻¹, indicating that the OH bond is coordinated with the palladium. All of the absorption bands in the final catalyst shifted to lower frequencies, thus confirming that the Pd ion was placed on the magnetic composite.

Fig. 3 shows the thermogravimetric analysis of Fe₃O₄@sepiolite-Pd²⁺. This analysis was recorded by heating the sample at a rate of 10 °C min⁻¹. The endothermic peak at lower temperatures (lower than 110 °C) can be attributed to the desorption of physically absorbed water on the surface of sepiolite and Fe₃O₄. This analysis indicates that the 1.03% weight loss in the range of 250–310 °C can be ascribed to the water loss in this area [18]. Therefore, this composite is stable at 600 °C, and according to this analysis, the Fe₃O₄@sepiolite-Pd²⁺ composite is a high-temperature stable catalyst. The amount of metal supported onto the Fe₃O₄@sepiolite was measured by induced coupled plasma analysis, and the amount of Pd was 0.032 g per gram of catalyst (0.3 mmol g⁻¹).

The hysteresis loop for Fe₃O₄@sepiolite-Pd²⁺ in an applied magnetic field at ambient temperatures was studied with the field swept from −8000 to +8000 Oe (Fig. 4). Bare nanocrystals of Fe₃O₄ have the high saturation magnetization of 73.7 emu g⁻¹ at room temperature [19]. This magnetization decreases to 34 emu g⁻¹ in the Fe₃O₄@sepiolite-Pd²⁺ catalyst, which demonstrates that the nanoparticles exhibit high permeability in magnetization. Thus, it is sufficient for magnetic separation using a conventional magnet.

Fig. 5(a) and (b) shows the SEM and TEM images of the Fe₃O₄@sepiolite-Pd²⁺. SEM images confirm the presence of spherical nanoparticles with a rough surface in the 23–58 nm diameter range (Fig. 5(a)). Transmission electron microscopy (TEM) confirmed the formation of single-phase Fe₃O₄ nanoparticles with spherical morphology (Fig. 5(b)).

The XRD patterns of two different stepwise synthetic pathways for the preparation of the catalyst are shown in Fig. 6. As compared with sepiolite, the XRD pattern of Fe₃O₄@sepiolite Fig. 6(a) displays five peaks at 30.4°, 35.8°, 43.5°, 54°, 57.5°, and 63.1°, which can be assigned to the characteristic peaks of the cubic magnetite phase. Moreover, the characteristic peaks of sepiolite in the XRD pattern of Fe₃O₄@sepiolite weaken obviously, indicating that the structure of sepiolite may be partly destroyed, resulting in confusional crystal lattice [20].

The six peaks of Fe₃O₄ can be seen in the XRD pattern of the synthesized Fe₃O₄@sepiolite-Pd²⁺ (Fig. 6(b)). This shows that the prepared Fe₃O₄ nanoparticles in this stage are pure with a spinel structure, and the support of the Fe₃O₄@sepiolite surface with PdCl₂ does not result in a phase change of the magnetic nanoparticles. The XRD patterns of the recycled catalyst after two different reactions are shown in Fig. 7. The XRD patterns of the catalyst after the Suzuki reaction Fig. 7(a) show that the catalyst's structure did not change after the reaction, but the XRD patterns of the catalyst after nitro reduction reaction shown in Fig. 7(b) indicate new three peaks around 40.44, 46.54, and 63.11, which, respectively, relate to the (111), (200), and (220) reflections. These peaks confirm the Pd metal in zero oxidation state, indicating the reduction of Pd²⁺ in the presence of NaBH₄ as a reduction agent. This change in the nanocatalyst did not reduce its activity [21,22].

The EDX spectrum of the Fe₃O₄@sepiolite-Pd²⁺ catalyst is outlined in Fig. 8. Atoms related to the catalyst structure, such as Fe, Pd, Mg, Al, Si, and O, are seen in this spectrum. The analyses indicated that Pd was coated in the catalyst structure.

After characterization, the activity of the catalyst was tested in both the reduction of nitroarene compounds with NaBH₄ and the CC coupling reaction of phenylboronic acid and aryl halide. First, the reaction of the phenylboronic acid with iodobenzene under various conditions was examined to establish the optimal conditions. The results are summarized in Table 1.

The first reaction was performed under the following conditions: the mixture of phenylboronic acid (1 mmol, 0.122 g) and iodobenzene (1 mmol, 0.204 g) was reacted in the absence of catalyst and 2 mmol (0.276 g) potassium carbonate as base in EtOH/H₂O (8 mL, 1:1) at 45 °C under an air atmosphere. After 15 min of reaction time, no product was found. This result showed that the catalyst plays a vital role in this reaction (Table 1, entry 1). To improve yield, 5 mg of Fe₃O₄@sepiolite-Pd²⁺ was used. Yield was increased up to 43%. When the amount of catalyst was increased to 10 mg, the yield was increased up to 65% (Table 1, entry 2–3). The use of Fe₃O₄, sepiolite, and Fe₃O₄@sepiolite as catalyst caused the efficiency of the reaction to decrease to 8%, 0%, and 5%, respectively (Table 1, entries 4–6). The effect of temperature and solvent on the reaction efficiency was also checked. Interestingly, the yield was increased up to 95% when the temperature was increased to 70 °C (Table 1, entry 7). The maximum yield (95%) was obtained when the mixture of EtOH/H₂O (2:1) and (1:1) was used as a solvent; thus, the eco-friendly and greener mixture EtOH/H₂O (1:1) was chosen as the optimal solvent (Table 1, entries 8–10). In addition, the role of different bases was examined; the use of KOH, Na₂CO₃, and CaCO₃ instead of K₂CO₃ reduced the yield to 12–88% (Table 1, entries 11–13).

With the optimal conditions determined (Scheme 2), the substrate scope of this reaction was next studied by screening of the aryl halides. A variety of aryl halides were subjected to optimal conditions, and the results are shown in Scheme 2.

All biphenyls products were known and characterized by melting point (mp), IR, and ¹H NMR data as well as by the comparison of their physical data with those in the literature [23–29].

The reduction of nitroarene compounds with sodium borohydride is a powerful synthetic method for the generation of aniline derivatives. In the next section, the catalytic activity of Fe₃O₄@sepiolite-Pd²⁺ for the reduction of nitroarene compounds is explored. First, the reaction of 2-nitroaniline with NaBH₄ under various conditions was examined to establish the optimal conditions. The results are summarized in Table 2.

The first reaction was performed under the following conditions: 2-nitroaniline (1 mmol), NaBH₄ as reducing agent (5 mmol), solvent (4 mL), and in the absence of any catalyst at 50 °C under an air atmosphere. After 50 min of reaction time, trace phenylenediamine was found. This result showed that the catalyst played a vital role in this reaction (Table 2, entry 1). To improve yield, 10 mg of Fe₃O₄@sepiolite-Pd²⁺ was used, and the yield was increased up to 45% (Table 2, entry 2). The addition of 10 mg of Fe₃O₄, sepiolite, and Fe₃O₄@sepiolite to the reaction mixture led to only traces of the product, and no catalytic activity for this reaction (Table 2, entries 3–5) took place. In an attempt to improve yield, the effects of both catalyst loading and temperature on the efficiency of the reaction were considered. Interestingly, the yield was increased up to 60% when 20 mg of the catalyst was used (Table 2, entry 6) and up to 96% when the temperature rose to 70 °C (Table 2, entry 7). However, the yield was not changed when an equimolar ratio of EtOH-H₂O (Table 2, entry 8) was used. When the reductant reagent was decreased to 3 mmol, the reaction yield was also reduced to 65% (Table 2, entry 9).

With the optimal conditions determined (Scheme 3), the substrate scope of this reaction was next studied by screening the nitroarene derivatives. A variety of nitroarenes were subjected to optimal conditions, and the results are depicted in Scheme 3.

All of the nitroarenes were converted to the corresponding anilines in high yields. The corresponding products were known and characterized by IR and ¹H NMR data as well as by the comparison of their physical data with those in the literature [30–34].

The mechanism of the reaction is not entirely clear, but a reasonable mechanism has been depicted in Scheme 4. On the basis of the reaction conditions, the first step of the mechanism was an oxidative addition; Ar-X attached to the catalyst, and an organopalladium compound was formed through a Pd(II)–Pd(IV) catalytic cycle [35]. The next step was transmetalation, and the Ph group was transferred from the activated phenylboronic acid to the palladium catalyst. In the final step of reductive elimination, biaryl was formed and the catalyst was regenerated (Scheme 4).

The efficacy of the Fe₃O₄@sepiolite-Pd²⁺ nanoparticles was compared with some previously reported catalysts through the reaction of bromobenzene with phenylboronic acid (see Table 3). The advantages of the catalyst introduced here are a short reaction time, easy separation of the catalyst, and its reusability.

The reusability of the catalyst was tested in the model reaction. After completion of the reaction in the first run, the catalyst was easily separated from the reaction mixture by using an external magnet (Fig. 9(b) and (c)). It was washed twice with ethyl acetate, dried at ambient temperature, and immediately used in the next step. The reaction was repeated for up to five consecutive runs with no significant change in the efficiency of the reaction observed (Fig. 9(a)).