Introduction
Acetalization reactions are widely used in organic synthesis for the purpose of protecting the carbonyl groups of aldehydes/ketones and synthesizing acetals which are important intermediates in the production of steroids, pharmaceuticals and fragrances [ 1]. Acetalization reactions have often been carried out using homogeneous acid catalysts such as H2SO4, HCl, and p-toluenesulfonic acid [ 1]. However, these homogeneous catalysts inevitably produce problems such as the need for the tedious purification of the products, the generation of a large amount of acidic waste, the corrosion of the equipment, and the production of severe environmental pollution. In order to overcome these disadvantages, much effort has been devoted to the development of heterogeneous solid acidic catalysts for acetalization reactions [ 2‒ 8].
Metal-organic frameworks (MOFs) are a burgeoning class of hybrid porous materials synthesized by the controlled assembly of metal ions/clusters with organic linkers. Typically, MOFs have unique features such as well-defined crystalline structures, large specific surface areas, and uniform and tunable cavities [ 9‒ 11]. These characteristics make them potentially useful in the realm of heterogeneous catalysis [ 10‒ 12]. MIL-100(Fe) (where MIL stands for Materials of Institute Lavoisier) is a representative carboxylate-based MOF which features a large surface area and a high resistance toward a variety of organic solvents. In addition, when it is dehydrated at temperatures up to 100 °C, MIL-100(Fe) possesses coordinatively unsaturated metal sites. Therefore, MIL-100(Fe) has been used as an effective Lewis acid catalyst in some acid-catalyzed organic synthesis reactions [ 13‒ 17].
The efficient separation and recycling of solid catalysts after catalytic reactions is very important for practical application [ 18‒ 20]. However, it is still a challenge to fabricate multifunctional recyclable MOFs-based materials, which can be applied in heterogeneous catalysis [ 21]. Recently, it was reported that the incorporation of magnetic iron oxide particles into porous MOFs does not alter the intrinsic properties of the pristine MOFs [ 22]. However, this modification does make it possible to quickly recycle the MOFs-based catalyst from a liquid system using a magnet.
Based on this work, a facile and convenient method for the preparation of a magnetically recyclable MIL-100(Fe)-based catalyst is reported herein. The MIL-100(Fe) was encapsulated on the outer surface of magnetic SiO2@Fe3O4 microspheres through an in-situ self-assembly method. The synthesized MIL-100(Fe)@SiO2@Fe3O4 was systematically characterized by XRD, N2 adsorption, FTIR, pyridine adsorption FTIR, SEM and TEM techniques. Additionally, the recyclability of MIL-100(Fe)@SiO2@Fe3O4 as a magnetically recyclable solid acidic catalyst was tested for the catalytic acetalization of aldehydes and diols.
Experimental
Synthesis of catalysts
The magnetic SiO2@Fe3O4 particles were prepared according to the procedure reported in the literature [ 22, 23]. The MIL-100(Fe)@SiO2@Fe3O4 microspheres were then fabricated by mixing 0.1 g of SiO2@Fe3O4 microspheres with Fe(NO3)3•9H2O (1.01 g, 2.5 mmol), H3BTC (0.47 g, 2.25 mmol) and deionized water (6 mL) in a flask under reflux (about 100 °C) for 8 h. The whole process is shown in Scheme 1. The obtained precipitate was rinsed three times with a solvent extraction treatment by using deionized water (30 mL) and ethanol (30 mL) at 70 °C for 24 h. Then the product was dried in vacuum at 150 °C for 10 h. Pure MIL-100(Fe) was synthesized by the same method but no SiO2@Fe3O4 particles were added.
Characterization
The powder X-ray diffraction (XRD) patterns of the samples were obtained on a Philips PW3040/60 diffractometer using CuKa radiation (40 kV, 30 mA, 0.1541 nm). The N2 adsorption isotherms were measured on a Micromeritics ASAP 2020 apparatus at ‒196 °C. Prior to the adsorption measurements, the samples were out-gassed under vacuum at 150 °C for 12 h. The scanning electron microscopy (SEM) observations were performed on a Hitachi S-4800 apparatus equipped with a field emission gun. Transmission electron microscopy (TEM) observations were carried out on a 2100 JEOL TEM (JEOL Ltd., Akishima, Tokyo, Japan) at 200 kV. The magnetic measurements were carried out on a superconducting quantum interference device (SQUID, MPMS, Quantum Design, San Diego, CA, USA). The FTIR spectra were collected on a Nicolet NEXUS670 Fourier transform IR spectrophotometer in KBr disks at room temperature. The numbers of acid sites in the MIL-100(Fe) and MIL-100(Fe)@SiO2@Fe3O4 samples were determined using pyridine (Py-FTIR), according to a previously reported method [ 17]. The acid capacity of Amberlyst-15 was determined by acid-base titration using a NaCl solution as an ion-exchange agent [ 17]. The acid amount of ultra-stable Y (USY) and H-beta zeolites were detected by ammonia temperature-programmed desorption (NH3-TPD) [ 17].
Acetalization of benzaldehyde and glycol
Benzaldehyde (35 mmol), glycol (64 mmol), and the catalyst (0.06 g) were added to cyclohexane (10 mL). The reaction mixture was maintained at 80 °C with a mechanical stirring speed of 960 r/min. The reaction mixture was analyzed by a gas chromatograph (Agilent 6820) equipped with an FID detector and a capillary column (DB-5, 30 m × 0.45 mm × 0.42 µm). For the recyclability tests, the catalyst was separated from the reaction mixture using an external magnet at the end of reaction. The catalyst was then thoroughly washed with acetone, treated at 150 °C for 3 h for reactivation and then used for another acetalization reaction.
Results and discussion
Figure 1 shows the SEM and TEM images of the Fe3O4, SiO2@Fe3O4 and MIL-100(Fe)@SiO2@Fe3O4 samples. For Fe3O4, the SEM image shows that the particles are nearly spherical in shape with rough surfaces (Fig. 1(A)). The TEM image confirms that the particles are uniform in size with diameters of ~350 nm (Fig. 1(C) and Fig. S1 in Electronic Supplemental Material). The surface roughness of the Fe3O4 particles can be attributed to the fact that the particles are formed by packing many tiny nanocrystals together [ 23, 24].
The SEM (Fig. 1(B)) and TEM (Fig. 1(D)) images for SiO2@Fe3O4 also show uniform particles although the average diameters have increased to about 360 nm. During the hydrolysis of TEOS, the Fe3O4 becomes coated with a compact layer of SiO2 with a thickness of about 10 nm. The silica coating is important to prevent the magnetite particles from leaching under harsh conditions [ 25].
A TEM image of a MIL-100(Fe)@SiO2@Fe3O4 microsphere is shown in Fig. 1(E). These particles were formed by wrapping the uniform SiO2@Fe3O4 spheres with a layer of highly stable metal-organic MIL-100(Fe) framework via the in-situ self-assembly of Fe(NO3)3 and H3BTC in aqueous solution at elevated temperature [ 17]. The TEM image shows that the SiO2@Fe3O4 microsphere is well encapsulated in the MIL-100(Fe) and that a core-shell structure was formed. The mean thickness of the MIL-100(Fe) shell is about 80 nm.
To further verify the structure of MIL-100(Fe)@SiO2@Fe3O4, XRD patterns of Fe3O4, SiO2@Fe3O4 and MIL-100(Fe)@SiO2@Fe3O4 were collected and the results are shown in Fig. 2. The diffraction peaks in Fe3O4 can be indexed to the face-centered cubic lattice of Fe3O4 spheres according to the Joint Committee on Powder Diffraction Standards (JCPDS) 75-1609 [ 23, 24]. The diffraction peaks of SiO2@Fe3O4 are almost the same as those for Fe3O4 although some of the intensities are different which is probably due to the introduction of the amorphous SiO2 layer on the outer surface of Fe3O4. The diffraction pattern of MIL-100(Fe)@SiO2@Fe3O4 contains the Fe3O4 reflections as well as some new diffraction peaks. These peaks likely correspond to the added MOF MIL-100(Fe) layer, since the XRD curve of MIL-100(Fe) synthesized in the current case is very similar with the simulated one that derived from the crystal structure data [ 13, 14]. All peaks in the MIL-100(Fe)@SiO2@Fe3O4 curve can be well ascribed to the diffraction peaks of MIL-100(Fe) and Fe3O4. These results demonstrate that MIL-100(Fe)@SiO2@Fe3O4 was successfully synthesized.
Figure 3(A) shows the nitrogen adsorption isotherms of MIL-100(Fe) and MIL-100(Fe)@SiO2@Fe3O4, from which the Brunauer-Emmett-Teller (BET) specific surface area (SBET) and pore volume (Vtotal) can be determined. These values for both materials are given in Table 1. In comparison with the pristine MIL-100(Fe), both of the SBET and Vtotal of MIL-100(Fe)@SiO2@Fe3O4 decrease significantly which is probably mostly due to the heavier and nonporous cores in MIL-100(Fe)@SiO2@Fe3O4 [ 20]. The pore size distribution of both MIL-100(Fe) and MIL-100(Fe)@SiO2@Fe3O4 are shown in Fig. 3(B). Both materials have two different pore sizes centered at about 1.8 and 2.2 nm. This distribution is consistent with previous reports of pure MIL-100(Fe) [ 17].
Tab.1 Physicochemical properties of various catalysts |
| Catalyst | SBETa /(m2·g-1) | Vmicrob /(cm3·g-1) | Vmeso /(cm3·g-1) | Acid amount /(mmol·g-1) |
|---|---|---|---|---|
| MIL-100(Fe) | 1836 | 0.34 | 0.82 | 1.92c |
| MIL-100(Fe)@SiO2@Fe3O4 | 340.7 | 0.08 | 0.22 | 0.45c |
| USY | 595 | 0.24 | 0.072 | 0.95d |
| H-Beta | 518 | 0.20 | 0.059 | 1.17d |
| SiO2@Fe3O4 | 2.6 | 0 | 0 | 0.01d |
| Amberlyst-15 | 38 | 0 | 0.18 | 4.7e |
| MIL-100(Fe)@SiO2@Fe3O4 | 329.7 | 0.07 | 0.21 | 0.41c |
Figure 4 shows the FTIR spectra of Fe3O4, SiO2@Fe3O4, MIL-100(Fe), and MIL-100(Fe)@SiO2@Fe3O4. Compared to the spectra for Fe3O4 and SiO2@Fe3O4, the spectrum of MIL-100(Fe)@SiO2@Fe3O4 has additional adsorption bands which have been previously shown to be associated with the MIL-100(Fe) structure [ 17]. For example, the strong vibration at 1628 cm-1 can be attributed to the interaction between the deprotonated –COOH and the Fe ion [ 14]. The vibration at 1385 cm-1 is mainly due to symmetric –COOH stretching. The appearance of these characteristic MIL-100(Fe) bands indicates the growth of MIL-100(Fe) crystals on the surface of the SiO2@Fe3O4 spheres.
The magnetic properties of the samples are shown in Fig. 5. The magnetic saturation (MS) values for Fe3O4 and the MIL-100(Fe)@SiO2@Fe3O4 core-shell particles are about 82.5 and 30.0 emu•g-1, respectively. The significantly lower MS value of the core-shell catalyst is due to the dielectric property of the MIL-100(Fe) and SiO2 shells [ 20]. Magnetic-saturation implies that the material should have a strong magnetic responsivity. The magnetic MIL-100(Fe)@SiO2@Fe3O4 core-shell particles are highly dispersible in solution and remain dispersed for more than ten hours before precipitating from the solution. This good dispersiblity is due to their submicron particle sizes. However, the particles also exhibit a rapid response to an external magnetic field. A permanent magnet (14000 Gs) placed near the reactor can remove the MIL-100(Fe)@SiO2@Fe3O4 core-shell particles from the solution within 10 s (inset Fig. 5).
The catalytic activity of the magnetic MIL-100(Fe)@SiO2@Fe3O4 hybrid for the liquid-phase acetalization of benzaldehyde with glycol was then tested. First, the amount of catalyst, the molar ratio of benzaldehyde to glycol, the reaction time, and the temperature on the catalytic performance were optimized (Fig. S2 in Electronic Supplemental Material). The optimal reaction conditions for the conversion of benzaldehyde were 0.06 g of catalyst, a benzaldehyde to glycol molar ratio of 0.55, and 80 °C. The amount of benzaladehyde that was converted with the MIL-100(Fe)@SiO2@Fe3O4 catalyst as a function of reaction time is shown in Fig. 6. An activity of 73.2% conversion was achieved after 120 min of reaction at 80 °C. In contrast, SiO2@Fe3O4 gave a very low conversion of benzaldehyde (14.9%) after 120 min (Fig. 6). This can probably be ascribed to the limited number of acid sites in SiO2@Fe3O4 (Table 1). These results suggest that an acidic catalyst is essential for the acetalization reaction. The high catalytic activity of MIL-100(Fe)@SiO2@Fe3O4 is probably due to the large number of acid sites in MIL-100(Fe) (Table 1).
Fig.7 Acetalization of benzaldehyde and glycol in the presence of MIL-100(Fe)@SiO2@Fe3O4 and SiO2@Fe3O4 as a function of reaction time. Reaction conditions: 0.06 g of catalyst, 35 mmol of benzaldehyde, 64 mmol of glycol and 10 mL of cyclohexane used as the water removal agent at 80 °C |
A leaching experiment was also carried out in order to confirm the key role of MIL-100(Fe)@SiO2@Fe3O4 as the catalytically active species as well as to confirm the heterogeneous nature of the catalyst. Upon removal of the catalyst, the mixture was allowed to react in the absence of the catalyst for 120 min (Fig. 6). There was no observable increase in the conversion, which strongly suggests that the acetalization reaction was heterogeneously catalyzed. This observation is consistent with the Inductively Coupled Plasma Optical Emission Spectrometry (ICP-OES) analysis of the above reaction mixture which showed no traces of Fe (below the detection limit, 0.5 ppm). Additionally, the specific surface area, the pore volume and the number of acidic sites in the used catalyst are very close to those in the fresh catalyst (see Table 1). All these results show that MIL-100(Fe)@SiO2@Fe3O4 is a highly active and stable solid catalyst for liquid-phase acetalizations.
Considering that the USY and H-Beta are important zeolites catalysts [ 7], and Amberlyst-15 is a typically commercial acidic resin catalyst [ 14], therefore, their catalytic activities in the acetalization were also tested and compared with those of MIL-100(Fe) and MIL-100(Fe)@SiO2@Fe3O4, and the results are shown in Fig. 7. Among the catalysts investigated, the fresh Amberlyst-15 showed a highest activity on a weight basis. The high catalytic activity of Amberlyst-15 can probably be attributed to its high number of acid sites available (4.7 mmol·g-1, Table 1), which is much higher than those of the other catalysts. In contrast, the USY and H-Beta catalysts showed comparably lower catalytic activities which is probably due to intracrystalline diffusion limitations [ 8]. On the other hand, it is interesting to noted that catalytic activity over MIL-100(Fe)@SiO2@Fe3O4 was almost the same with that of MIL-100(Fe). Considering that the low density of the parent MIL-100(Fe) (close to 1 g/cm3) [ 13], easily resulting in inconvenient separation from liquid medium when used as a heterogeneous catalyst. Thus, the developed MIL-100(Fe)@SiO2@Fe3O4 may have the advantages of convenient recycling and reuse.
Fig.8 Activity comparison of different catalysts in the acetalization of benzaldehyde with glycol. (a) blank; (b) MIL-100(Fe)@SiO2@Fe3O4; (c) MIL-100(Fe); (d) USY; (e) H-Beta; (f) Amberlyst-15. Reaction conditions: 0.06 g of catalyst, 35 mmol of benzaldehyde, 64 mmol of glycol and 10 mL of cyclohexane used as the water removal agent at 80 °C for 120 min |
Next the recyclability of the MIL-100(Fe)@SiO2@Fe3O4 was tested. After the reaction, a magnet was placed close to the reactor wall in order to remove the magnetic MIL-100(Fe)@SiO2@Fe3O4 catalyst from the reaction mixture (Fig. 5). The recovered catalyst was then reused for another seven experiments under identical reaction conditions and the results are shown in Fig. 8. The recovered MIL-100(Fe)@SiO2@Fe3O4 catalyst was weighed and at least 95 wt-% was recovered. The activity of the reused MIL-100(Fe)@SiO2@Fe3O4 catalyst decreased slightly with each reuse (from 73.2 to 66.9%) (Fig. 8(A)). In another experiment, a small amount of fresh catalyst was added to the recovered catalyst in order to give a constant amount of catalyst (0.06 g). As shown in Fig. 8(A), in this case there was no loss of benzaldehyde conversion with repeated usage. These results combined with the above discussion on the nature of the MIL-100(Fe)@SiO2@Fe3O4 catalyst lead to the conclusion that the decrease in the activity of the recovered catalyst is due to the slight loss of catalyst that occurs during the recovery and transfer processes from one run to the next.
Fig.9 Reusability of (A) MIL-100(Fe)@SiO2@Fe3O4 and (B) Amberlyst-15 in the catalytic acetalization of aldehyde with glycol. Reaction conditions: 0.06 g of catalyst, 35 mmol of benzaldehyde, 64 mmol of glycol and 10 mL of cyclohexane used as the water removal agent at 80 °C for 120 min. (a) recovered catalyst with no fresh catalyst added and (b) fresh catalyst added to maintain a constant amount (0.06 g) of catalyst |
The reusability of Amberlyst-15 was tested and the results are shown in Fig. 8(B). The catalytic activity of Amberlyst-15 decreased substantially from the first to the eighth run. The recovery ratio for Amberlyst-15 was about 96 wt-%. When fresh catalyst was added to the used Amberlyst-15 to give 0.06 g of catalyst, the catalytic activity improved but the overall activity still gradually decreased during repeated usage. This is probably the result of catalyst swelling [ 17].
Conclusions
A magnetic MIL-100(Fe)@SiO2@Fe3O4 catalyst with a novel core-shell structure was prepared via a facile in-situ self-assembly method. It should be stated that Fe3O4@MIL-100(Fe) hybrids have been previously prepared by Qiu and coworkers following a so-called step-by-step strategy in the presence of an organic solvent [ 24‒ 26]. In their method multiple steps were required, and the synthesis typically involved at least 40 repetitions with each step taking 30 min. In contrast, our synthesis was conducted in aqueous media, which is more environmentally friendly. In addition, the MIL-100(Fe)@SiO2@Fe3O4 microspheres were fabricated within 8 h in one step by the direct mixing of SiO2@Fe3O4 with the MIL-100(Fe) precursors resulting is a huge time savings. This catalyst has good catalytic activity for the acetalization of benzaldehyde and glycol which is due to the large number of Lewis acid sites derived from the MIL-100(Fe). Importantly, the MIL-100(Fe)@SiO2@Fe3O4 catalyst has superparamagnetic properties and can be easily separated from the solid-liquid reaction system using an external magnet. Thus, a magnetically controllable on-off reaction was demonstrated for the catalytic acetalization over the developed MIL-100(Fe)@SiO2@Fe3O4.