A magnetically recyclable superparamagnetic silica supported Pt nanocatalyst through a multi-carboxyl linker: synthesis, characterization, and applications in alkene hydrosilylation

Abstract

To simplify separation procedures, improve the reusability and decrease the loss of Pt, two Pt catalysts anchored on superparamagnetic silica (Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt) were prepared for the first time. The stable magnetic properties made them easily recyclable using a magnet rather than filtration, decantation or centrifugation. After 12 catalytic runs for both 30–50 nm Pt catalysts, the yield of 1-heptylmethyldichlorosilane was still up to 90%. The average loss of Pt in each reaction was only 0.87% for Fe₃O₄@SiO₂-EDTA@Pt and 0.66% for Fe₃O₄@SiO₂-DTPA@Pt owing to the strong interaction between Pt and carboxyl. The unprecedented activity and selectivity of the two Pt nanoparticle catalysts were observed in the hydrosilylation of alkenes. The turnover number in the reaction between 1-hexene and methyldichlorosilane using 5 × 10⁻⁸ mol of the Pt approached 662 733 for Fe₃O₄@SiO₂-EDTA@Pt and 579 947 for Fe₃O₄@SiO₂-DTPA@Pt over 12 h. The corresponding hydrosilylation products in excellent yields were obtained when we employed a broad range of alkenes as substrates, including 5 isomerous hexenes and 14 important industry raw materials. Fe₃O₄@SiO₂-DTPA@Pt showed a better activity. They have potential for catalyzing more reactions and replacing the current homogeneous Pt catalysts in industry.

---

1. Introduction

Magnetic materials such as magnetite (Fe₃O₄) and maghemite (γ-Fe₂O₃) are widely utilized in various fields, such as high-density data storage, chemical sensors, catalysts, drug delivery, biological assays, and electro photographic developers.^1–5 Compared with traditional techniques such as filtration, decantation or centrifugation, magnetic materials could be easily separated from solution and re-dispersed through an external magnetic field,⁶ which overcomes time- and solvent-consuming procedures.⁶ Magnetite Fe₃O₄, especially nano-Fe₃O₄ possessing super paramagnetic behavior at a nanoparticle size with high surface area, is attractive for practical applications.^7–10 An ongoing application of magnetic materials is to immobilize metal catalysts because efficient separation, recovery and recycling of the catalyst may enhance their lifetime and minimize their consumption to result in significant economical and environmental benefits.^11–14

Among these transition metal catalysts, such as copper (Cu), platinum (Pt), palladium (Pd), rhodium (Rh), nickel (Ni) or ruthenium (Ru),^15–20 Pt is one of the most widely used owing to its unparalleled catalytic activity and mild reaction conditions in various fields, such as hydrosilylation, aldehyde oxidation, sulfuric acid decomposition. Although homogeneous Pt catalysts such as Speier's catalyst and the Karstedt's catalyst suffer from a number of drawbacks, like side reactions and low selectivity and reusability, there isn't efficient substitution in industry. The immobilization of platinum catalyst is a wise choice. Numerous researches were being reported on how to immobilize Pt on magnetic materials.^21–26

Magnetic silica is used as an ideal inorganic support because of its merits, like magnetic property, easy modification, high biocompatibility, solvent compatibility, adsorption capacity, acid–base properties, and chemical and thermal stability when compared to other structures.^27–29 A number of multi-carboxyl based affinity ligands, such as ethylenediaminetetraacetic acid (EDTA) and diethylenetriaminepentaacetic acid (DTPA), are well-known chelators for their strong metal-complexing property.³⁰ Magnetic silica particles modified by EDTA or DTPA were widely used in industries including cleaning, the nuclear industry, pharmaceuticals, and the manufacturing of textile leather, rubber, and paper, etc.^31–35 The immobilized Pt catalysts on silica gel through EDTA and DTPA were developed by our group.^36,37 EDTA and DTPA showed a strong interaction with platinum, but the kinds of heterogeneous catalysts still need to reuse through centrifugation and 31% Pt was loss after 13 recycles.

Here, we first time try to immobilize Pt catalysts through carboxyl ligands like EDTA or DTPA grafted onto the surface of magnetite silica (Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt) to simplify separation procedure, improve the reusability and decrease the loss of Pt. In addition, the two catalysts are applied first time in 4 isomerization–hydrosilylation and hydrosilylation of 14 compounds, which are important industry raw materials. These make the current two catalysts more valuable industry.

2. Experimental section

2.1. Materials and reagents

All chemicals were used as received without further purification. Iron chloride hexahydrate (FeCl₃·6H₂O, >99%) and ferrous chloride tetrahydrate (FeCl₂·4H₂O, >99%) were from Yuanli Chemical. γ-Aminopropyltrimethoxysilane (APTMS) was from Qufu Chenguang Chemical. Tetraethylorthosilicate (TEOS) was from Tianjin Kemiou Chemical Reagent. Ethylenediaminetetraacetic acid (EDTA), diethylenetriaminepentaacetic acid (DTPA), sodium silicate, ammonia solution (25%), anhydrous pyridine, acetic anhydride, diethyl ether, hydrochloric acid (HCl), toluene, acetone, ethanol, methanol, n-decane, i-propanol, n-butanol and n-hexanol were from Jiangtian Chemical Technology. Methyldichlorosilane was from Kaihua Synthetic. 1-Hexene, trans-2-hexene, cis-2-hexene, trans-3-hexene and cis-3-hexene were form Tokyo Chemical Industry Co., Ltd. N,N-Dimethylacrylamide, 1,2-epoxy-4-vinylcyclohexane, methyl methacrylate, butyl acrylate, ethane-1,2-diyl bis(2-methylacrylate), crotonic acid ethyl ester and trans-2-hexenyl acetate were from Bid Pharmatech Ltd. 1,2-Epoxy-7-octene, allyl glycidyl ether, 5-hexen-2-one, allylchloride and trans-2-hexen-1-yl phenylacetate were from Nanjin Shengbicheng Chemical Technology Co., Ltd. Cyclohexene and norbornylene were from Shanghai Macklin Biochemical Co., Ltd. H₂PtCl₆·6H₂O was from Tianjin Fengchuan Chemical Reagent Technologies. 1-Hexyl-methyldichlorosilane was from CNW Technologies. Deionized water was from Tianjin Yongyuan Distilled Water Manufacturing Center.

2.2. Materials synthesis

2.2.1. Preparation of Fe₃O₄ and Fe₃O₄@SiO₂. The magnetite Fe₃O₄ and Fe₃O₄@SiO₂ nanoparticles were prepared according to previous reported method with a minor modification.^38–43 The magnetite Fe₃O₄ nanoparticles were prepared by chemical co-precipitation of ferric and ferrous ions in ammonium hydroxide solution. Briefly, under nitrogen protection, 0.04 mol FeCl₃·6H₂O and 0.02 mol FeCl₂·4H₂O were dissolved in distilled water with vigorous stirring at 80 °C for 20 min. After 15 mL ammonium hydroxide (25% aqueous solution) was added to the solution, the color of the bulk solution turned from orange to black immediately. After 30 min of reaction, the black precipitate was collected from the reaction mixture by a powerful magnet and washed three times with distilled water before it was re-dispersed into a 100 mL distilled water to form a nanoparticle suspension of ca. 50 mg mL⁻¹.

The magnetite Fe₃O₄@SiO₂ nanoparticles were prepared by two sol–gel approach. First, the surface of Fe₃O₄ nanoparticles was covered by silica layer (SiO₂ encapsulated Fe₃O₄ particles, Fe₃O₄@SiO₂). Briefly, sodium silicate was dissolved in deionized water in a glass beaker equipped with a mechanical stirrer and the pH was adjusted to 12–13 by sodium hydroxide. Fe₃O₄ nanoparticles were added to the sodium silicate solution. The mixture was treated ultrasonically for 30 min. Then, the temperature of the mixture was increased to 85 °C and nitrogen gas was used to prevent the intrusion of oxygen. 2 M hydrochloric acid was added dropwise to adjust pH value to 6.0. Then, the precipitates were washed several times with deionized water by magnetic decantation. The Fe₃O₄@SiO₂ was coated again to ensure that the desired amount of silica on all magnetite Fe₃O₄.^44,45 For this, Fe₃O₄@SiO₂ was re-dispersed in 100 mL methanol–water (1 : 1, v/v) mixture and followed by the addition of 1 mL ammonium hydroxide (25% aqueous solution). Then, 1 mL of TEOS was dissolved in 10 mL methanol and added dropwise to the mixture by gently stirring for 1 h. The obtained nanoparticles (Fe₃O₄@SiO₂) were collected by magnet and washed several times with distilled water and ethanol. Finally, the nanoparticles were dried at room temperature under vacuum and brown yellow solid was observed.

2.2.2. Preparation of amino-magnetic silica (Fe₃O₄@SiO₂-AP). It was prepared according to previous reported method.⁴⁶ Ten grams as-made magnetic silica nanoparticles were ultrasonically dispersed in 100 mL distilled toluene for 30 min. Then, it was transferred into a 250 mL three-necked flask equipped with a mechanical stirrer. 3-Aminopropyltrimethoxysilane (20 mL) was added into this mixture. The temperature was kept at 110 °C with continuous stirring for 24 h. The resulting nanoparticles were separated by magnet, washed with methanol for several times and then dried at 60 °C under vacuum for 12 h. The product was referred to as Fe₃O₄@SiO₂-AP.

2.2.3. Preparation of multi-carboxyl magnetic silica (Fe₃O₄@SiO₂-EDTA and Fe₃O₄@SiO₂-DTPA). The preparation of EDTA or DTPA dianhydride was done according to a previous reported method with minor modification.⁴⁷ EDTA or DTPA was suspended in pyridine followed by the addition of acetic anhydride. The mixture was stirred at 65 °C for 24 h. The product was filtered, washed with acetic anhydride and diethyl ether, and dried in an oven for 6 h. Next, 5 g of Fe₃O₄@SiO₂-AP was suspended in 100 mL mixture solution of ethanol and acetic acid (1 : 1, v/v). After the addition of 25 g of EDTA dianhydride or 30 g DTPA dianhydride, the mixture was kept at 80 °C for 24 h to allow the amino groups on the silica gel to react with the carboxyl groups.^48,49 Finally the yellowish-brown product was separated by magnet and washed with 4 M ammonia water solution to remove unreacted EDTA or DTPA.⁵⁰ Then products were washed again with ethanol/acetic acid (1 : 1, v/v), water and methanol sequentially.⁵¹ The multi-carboxyl modified magnetic silica gels (Fe₃O₄@SiO₂-EDTA or Fe₃O₄@SiO₂-DTPA) were dried in an oven at 90 °C for 8 h.

2.2.4. Preparation of catalysts. The above multi-carboxyl modified magnetic silica gels (Fe₃O₄@SiO₂-EDTA or Fe₃O₄@SiO₂-DTPA) (1.51 g) were added to 50 mL H₂PtCl₆·6H₂O (0.20 g) ethanol solution. The mixture was carried out in a reflux at 80 °C under nitrogen atmosphere and stirred for 9 h. The solid product was filtered by a magnet and washed with ethanol successively and dried at 70 °C for 3 h. The light yellow platinum complex was obtained, which were denoted as Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt. The process was schematically illustrated in Fig. 1.

Fig. 1 Preparation of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt catalysts.

2.3. Characterization of the catalysts

The crystal lattice structure of the magnetic intermediates (Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP), and products (Fe₃O₄@SiO₂-EDTA@Pt, Fe₃O₄@SiO₂-DTPA@Pt) were identified by XRD, in the 2θ range of 10–90° at a scan speed of 4° min⁻¹, using a Rigaku D/MAX-2500 diffractometer with Cu-Kα radiation (λ = 0.1540 nm). Particle sizes of the catalysts were studied using a JEOL JEM-2100F apparatus with an accelerating voltage of 200 kV. TEM of Fe₃O₄@SiO₂-EDTA@Pt, Fe₃O₄@SiO₂-DTPA@Pt were prepared by placing a drop containing the nanoparticles in a coated carbon grid. The morphology of the catalysts was investigated using a Hitachi S2800 SEM. The EDS analysis, which was simultaneously performed during the SEM examinations, was conducted to confirm the element distribution on the surface of adsorbents. FT-IR spectra were taken in KBr pressed pellets on a TENSOR 27 Spectrometer. EA was used to measure the nitrogen content in the materials on an Elementar Vario Micro cube. The thermal stabilities were measured by TGA employing a NETZSCH STA 409 TG-DTA under a dynamic nitrogen atmosphere at a heating rate of 10 °C min⁻¹. The magnetization and hysteresis loops were measured at room temperature by a LDJ9600 VSM. The BET measurement was carried out on a Quantachrome NOVA 1000 system at liquid-N₂ temperature.

2.4. Catalytic alkene hydrosilylation

In a general procedure, hydrosilylation was carried out in a 50 mL round-bottomed tube equipped with a homemade reflux condenser to the upper of which a heating system was attached. An appropriate amount of alkene and platinum magnetic complex were stirred at a specific reaction temperature for 30 min before 15 mmol methyldichlorosilane was added. The structure and yield of hydrosilylation products were determined based on calibration curves of standards. The catalyst was collected by simple magnetic separation and reused in subsequent runs without any treatment. All hydrosilylation products were characterized by comparing their spectra and physical parameters with authentic samples. The reaction product was quantified by gas chromatography (GC) and the structures of products were identified by nuclear magnetic resonance (NMR). GC equipped with a flame ionization detector (FID) (Wuhan Trustworthy Technology, China) and a long capillary column (30 m × 0.25 mm × 0.25 μm) coated by 5% phenyl and 95% methyl polysiloxane. Injector and detector temperature was 260 °C, carrier gas was nitrogen, injection volume at 0.8 μL, temperature programme: the column was kept at 60 °C for 3 min, then heated to 260 °C at a rate of 10 °C min⁻¹. ¹H NMR spectra were recorded on a Bruker AC-P400 (400 MHz) spectrometer in CDCl₃ as solvent.

3. Results and discussion

3.1. Synthesis

In the synthetic procedures described earlier, magnetite Fe₃O₄ nanoparticles were prepared by the co-precipitation method from ferrous and ferric ion solutions at a mol ratio of 1 : 2. The magnetic nanoparticles have extensive hydroxyl groups on its surface when contacted with aqueous phase.⁵² Silicic acid as silica precursors encapsulated the Fe₃O₄ nanoparticles by two sol–gel approaches. First silicate acid layer was formed gradually from sodium silicate solution by dropwise addition of hydrochloric acid. The second silicate acid layer was formed by dropwise addition of TEOS solution. However, the TEOS is non-polar in nature and it is necessary to use organic reagent like methanol as a mutual solvent to make TEOS and water miscible. Subsequently, the silanol groups (Si–OH) on the surface are transformed into siloxane bonds (Si–O–Si) via the condensation reaction and the silica coating is formed.⁵³ Notably, the introduction of a dense silica coating on the magnetic nanoparticles is necessary for this study. It protects the magnetic Fe₃O₄ from oxidation and provides the surface charges needed to prevent aggregation in the subsequent condensation reactions and practical applications. The silanol groups change the surface properties making it easier and more effective to perform surface modification with organic silane molecules.

Before being modified by EDTA or DTPA group, the magnetic silica nanoparticles were functionalized with APTMS, which acted as an auto-catalyst. Those amino groups on the magnetic silica surface enable the magnetic silica materials to be functionalized with EDTA and DTPA by grafting the corresponding anhydrides on those amino groups. The EDTA or DTPA groups, in turn, serve as a new matrix to immobilize metallic Pt. The chemical structures of the resulting product were depicted in Fig. 1. The multi-carboxyl functionalized magnetic silica materials were found to be a yellow powder with an excellent shelf life. These magnetite nanoparticles functionalized with EDTA or DTPA will be referred to as Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt.

3.2. Characterization

The data from AAS analysis indicated the amounts of Pt in Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt were 173.80 and 192.77 μmol g⁻¹, respectively. As a comparison, Pt was also immobilized on Fe₃O₄@SiO₂-AP and Fe₃O₄@SiO₂, which had only 33.40 and 0.10 μmol g⁻¹ of Pt, respectively. It means the interaction between Pt and carboxyl is stronger than amino group, which is benefit to the immobilization and stability of Pt.

Morphology of the two novel catalysts was characterized by SEM, EDS and TEM (Fig. 2). Results showed that the dried power of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles had an irregular surface with sharp edge and slight agglomeration, which could be dispersed well in solution. The presence of Si, O and Fe signals in the EDS pattern confirmed that Fe₃O₄ was successfully mixed with silica. The higher intensity of Si peak compared with that of Fe peak showed that the Fe₃O₄ nanoparticles were coated by SiO₂. The collected process by a magnet avoids the SiO₂ nanoparticles without Fe₃O₄. The EDS analysis results also confirmed the elemental composition of the nanoparticles and the existence of Pt in the prepared nanoparticle catalysts. Through second robust sol–gel approach, the high magnification TEM images revealed that the Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt particles obtained were uniform with a diameter of about 30–50 nm. Each magnetic particle was composed of dark magnetic Fe₃O₄ nanoparticles of about 10–15 nm diameter. Even further, it clearly showed that those dark magnetic particles were coated with a light-gray SiO₂ layer with a thickness of around 15 nm.

Fig. 2 SEM micrographs of (a) Fe₃O₄@SiO₂-EDTA@Pt and (b) Fe₃O₄@SiO₂-DTPA@Pt, EDS of (c) Fe₃O₄@SiO₂-EDTA@Pt and (d) Fe₃O₄@SiO₂-DTPA@Pt, TEM micrographs of (e) Fe₃O₄@SiO₂-EDTA@Pt and (f) Fe₃O₄@SiO₂-DTPA@Pt.

The XRD patterns of Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles were shown in Fig. 3. It showed typical X-ray diffraction patterns of Fe₃O₄.⁵⁴ The characteristic diffraction peaks appeared in Fe₃O₄ at 30.1°, 35.5°, 43.3°, 54.3°, 57.3° and 62.6° could be assigned to the planes of Fe₃O₄ (JCPDS No. 19629), respectively. These characteristic diffraction peaks were observed in Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles which indicated that Fe₃O₄ was existent in all synthesized samples and the whole preparation processes did not result in the phase change of Fe₃O₄. The obvious characteristic peak of Pt wasn't observed in Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt, which may indicate that the amount of Pt was low or the Pt particles were highly dispersed on the surface of support.⁵⁵

Fig. 3 XRD patterns of Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles.

In order to confirm properties of the surface of composite nanoparticles, the molecular structure of the different composites was characterized using FT-IR spectra (Fig. 4.). The typical absorption peak for Fe₃O₄ at 635 and 576 cm⁻¹ was attributed to the stretching vibration of Fe–O bond. The typical peaks of Fe₃O₄ were observed in all of spectra, which further confirmed Fe₃O₄ nanoparticles existed in all of intermediates and products. Whereas the gradual decrease of peak signals indicated the Fe₃O₄ nanoparticles were at core of Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA, Fe₃O₄@SiO₂-EDTA@Pt, Fe₃O₄@SiO₂-DTPA and Fe₃O₄@SiO₂-DTPA@Pt. For Fe₃O₄@SiO₂, a new board absorption peak at 1092 cm⁻¹ resulted from Si–O–Si antisymmetric stretching vibration of silica layer, which verified the successful coating of silica layers onto Fe₃O₄.^56,57 The peak around 1627 cm⁻¹ was attributed to the absorbed water on the silica shell or the silanol group of the silica.⁵⁸ The characteristic absorption band at 3422 cm⁻¹ was from the stretching vibrations of hydroxyl groups. These peaks from Fe₃O₄@SiO₂ were also observed in Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA, Fe₃O₄@SiO₂-DTPA, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt. After the functionalization of Fe₃O₄@SiO₂ with APTMS, the new absorption peaks at 2925 and 2858 cm⁻¹ represented the symmetric vibrations of aliphatic C–H stretching of the methylene group for silane coupling agent, which indicated that amino group was successfully grafted onto the surface of Fe₃O₄@SiO₂. After grafting multi-carboxyl groups, the strong characteristic absorption band appeared at 1733 and 1404 cm⁻¹ could be assigned to C=O and C–O stretching vibration of the –COO– group, which indicated the formation of carboxylic groups (EDTA and DTPA) on the surface of the Fe₃O₄@SiO₂-AP.⁵⁹ As a result of the presence of large quantities of metallic Pt on the surface of the nanoparticles, the decrease in vibration of C=O and C–O suggests that the carboxyl group on the surface of the functional magnetic silica have coordinated with platinum, signifying the immobilization of platinum complex on the Fe₃O₄@SiO₂-EDTA and Fe₃O₄@SiO₂-DTPA surface.

Fig. 4 FT-IR spectra of Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA, Fe₃O₄@SiO₂-EDTA@Pt, Fe₃O₄@SiO₂-DTPA, and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles.

XPS was used to further characterize the above catalysts (Fig. 5.). XPS results showed distinct Fe, O, N, C, Si and Pt peaks from the two novel nanoparticles (Fig. 5a). The detail XPS data for Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt, Fe₃O₄@SiO₂-DTPA@Pt and H₂PtCl₆·6H₂O were listed in Table 1. It revealed that the binding energies of Si 2p and O 1s in Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt were similar to those of Fe₃O₄@SiO₂-AP. The banding energy of Cl 2p in Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt was similar to that of H₂PtCl₆·6H₂O. It was noteworthy that the banding energy of Pt 4f7/2 in Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt was 3–5 eV more than that in H₂PtCl₆·6H₂O. It could be summarized that electron donation from carboxyl group to Pt to form the coordination bond between carboxyl and Pt in Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt, which was also consistent with the FT-IR results. The Pt 4f XPS spectra contain two peaks corresponding to Pt 4f7/2 and 4f5/2 states from the spin-orbital splitting and each peak was de-convoluted with Pt(0), Pt(II) and Pt(IV).^60–62 In Fe₃O₄@SiO₂-EDTA@Pt, the percentage of Pt⁴⁺, Pt²⁺ and Pt⁰ peaks were 28.9, 46.5 and 24.6%, respectively. In Fe₃O₄@SiO₂-DTPA@Pt, the ratio of Pt²⁺ (50.3%) and Pt⁰ (26.4%) were higher than that of Fe₃O₄@SiO₂-EDTA@Pt, whereas the Pt⁴⁺ decreased to 23.3%. Results indicated the Pt²⁺ and Pt⁰ were predominant species that may be corresponding to good catalytic activity and might be produced on the catalyst surface through passivation process during the sample preparation.

Fig. 5 (a) XPS spectra of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt. (b) High-resolution XPS spectra of Pt 4f for Fe₃O₄@SiO₂-EDTA@Pt. (c) High-resolution XPS spectra of Pt 4f for Fe₃O₄@SiO₂-DTPA@Pt.

Table 1 XPS data for Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt, Fe₃O₄@SiO₂-AP and H₂PtCl₆·6H₂O (eV)^a

Samples	Pt 4f7/2	Si 2p	N 1s	O 1s	Cl 2p
a The banding energies are referred to C 1s (284.6 eV).
Fe3O4@SiO2-EDTA@Pt	78.40	107.21	405.59	536.88	201.29
Fe3O4@SiO2-DTPA@Pt	76.80	106.40	403.23	536.86	200.40
Fe3O4@SiO2-AP	—	108.78	407.24	537.60	—
H2PtCl6·6H2O	73.10	—	—	—	199.50

---

BET analysis showed another important characteristic of the Pt catalysts. Fig. 6a shows the nitrogen adsorption isotherms and pore-size distribution of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt catalysts. Both of two samples showed type III isotherms with type H3 hysteresis loops which is related to the slit-like pores because of aggregation of particles.⁶³ When we dispersed these nanoparticles in aqueous solution and observed using microscope, these nanoparticles were mono-disperse. It means the aggregation is temporary in BET analysis. The BET data including surface area, average pore size and total pore volume derived from the isotherms were calculated. The BET specific surface area, the pore size and the total pore volume of Fe₃O₄@SiO₂-EDTA@Pt were 75.68 m² g⁻¹, 17.50 nm and 0.685 cm³ g⁻¹, respectively. They were corresponding to 36.65 m² g⁻¹, 29.86 nm and 0.366 cm³ g⁻¹ for Fe₃O₄@SiO₂-DTPA@Pt.

Fig. 6 (a) Nitrogen adsorption–desorption isotherms and (inset) the corresponding pore-size distribution curves of Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt catalysts, (b) TGA curves of Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, SiO₂@Fe₃O₄-EDTA@Pt and SiO₂@Fe₃O₄-DTPA@Pt.

To test the thermal stability and the grafting amount of carboxyl groups on the magnetic silica nanoparticles, TGA analysis was performed for Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles (Fig. 6b.). As shown in the TGA curve of the prepared nanoparticles, the weight loss below 200 °C was attributed to the removal the water, and the organic parts were decomposed completely at about 700 °C. The TGA curve of bare Fe₃O₄@SiO₂ nanoparticles showed that the weight loss over a temperature range from 50 to 800 °C was about 4%, which was attributed to the escape of physically adsorbed water or/and structural water on the surface.⁶⁴ Moreover, the weight increases slightly in the temperature range of 700–800 °C, which might be caused by the slight oxidation of Fe₃O₄.⁶⁵ For Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt nanoparticles, the weight loss at temperatures below 200 °C was about 3%, which was related to the release of moisture and structural water on the surface of the silica layer. However, apart from the water loss, organic components inside the particles began to degrade rapidly at a temperature higher than 200 °C, and almost completely decomposed at 550 °C for Fe₃O₄@SiO₂-EDTA@Pt and 700 °C for Fe₃O₄@SiO₂-DTPA@Pt, respectively. Therefore, the weight loss from 200 to 800 °C in the TGA curves of Fe₃O₄@SiO₂-EDTA@Pt (about 28 wt%) and Fe₃O₄@SiO₂-DTPA@Pt (about 30 wt%) could be used to estimate the weight proportions of EDTA or DTPA-modified on Fe₃O₄@SiO₂. This phenomenon suggested that the weight proportion of multi-carboxyl groups on Fe₃O₄@SiO₂-DTPA@Pt (30 wt%) was higher than that on Fe₃O₄@SiO₂-EDTA@Pt (28 wt%). It was consistent with one more carboxyl group in DTPA than EDTA.

The key for magnetic materials is the sufficient magnetic property. Therefore the magnetic properties of these nanoparticles were analyzed by VSM (Fig. 7.). The saturation magnetization was found to be 74.7, 48.4, 44.4, 25.4 and 23.0 emu g⁻¹ for Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt, respectively. All of them showed superparamagnetic properties due to the presence of nanometer-sized magnetic Fe₃O₄ particles in the core.⁶⁶ Although the saturation magnetization values of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt were lower than that of Fe₃O₄, Fe₃O₄@SiO₂ and Fe₃O₄@SiO₂-AP, they could be readily separated from their dispersion in 1 min with an external magnetic field owing to their high magnetite content and thus good magnetic response (Fig. 7, inset). These results indicated that the magnetization of Fe₃O₄ decreases considerably after coated with SiO₂ due to the weight contribution from the nonmagnetic SiO₂. Nevertheless, both of the catalysts exhibited enough magnetic response to meet the need of magnetic separation.

Fig. 7 VSM magnetization curves of Fe₃O₄, Fe₃O₄@SiO₂, Fe₃O₄@SiO₂-AP, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt. Inset: magnetic separation of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt from the aqueous suspensions.

3.3. Catalyst performances for hydrosilylation

3.3.1 Activity and reusability in catalyzing hydrosilylation of 1-hexene. Hydrosilylation of alkene with silane is a typical reaction catalyzed by platinum, which exist two reaction pathways α-addition and β-addition.^67,68 Hydrosilylation of a series of alkenes with methyldichlorosilane were catalyzed by the two novel Pt catalysts, Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt, to evaluate their activity and regioselectivity.

Hydrosilylation conditions are important for yield and regioselectivity. To exhibit well the activities of two catalysts, we optimized hydrosilylation conditions of 1-hexene and methyldichlorosilane, including reaction temperatures (30, 40, 50, 60, 70, and 80 °C), times (0.5, 1, 2, 3, 4, and 5 h), the molar ratio of methyldichlorosilane and 1-hexene (0.5 : 1.0, 1.0 : 1.0, 1.2 : 1.0, 1.5 : 1.0, 1.8 : 1.0, and 2.0 : 1.0) and the additive sequence of 1-hexene and methyldichlorosilane (1-hexene added to methyldichlorosilane after methyldichlorosilane mixing with the catalyst for 30 min, simultaneous addition of 1-hexene and methyldichlorosilane, methyldichlorosilane added to 1-hexene after mixing with the catalyst for 30 min). The detail results were summarized in support information (Fig. 1S†). Results indicated both Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt (containing 2.01 × 10⁻³ mmol Pt in each catalyst) could efficiently catalyze the hydrosilylation. The yields were up to 99.4% and 98.77% when 19.2 mmol methyldichlorosilane was added to 11.8 mmol 1-hexene after mixing with the catalyst for 30 min and maintained the reaction 3 h at 50–60 °C. As a comparison, the yield only was 35.46% when Fe₃O₄@SiO₂@Pt was used as catalyst.

In research of platinum catalyst, the turnover number (TON) of Pt often is used as criteria. Thus, we calculated the TON approached 3628 and 3550 when Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt with 2.60 μmol Pt catalyzed 11.2 mmol of 1-hexene in 4 h. To further increase the TON for hydrosilylation, amount of Pt catalyst was gradually decreased to 2.10, 1.82, 1.47, 0.87, even 0.34 μmol. Results indicated the hydrosilylation between 1-hexene and methyldichlorosilane with low amount of Pt could be catalyzed and the TON quickly increased to 4721, 5435, 6744, 10 905 and 16 006 for Fe₃O₄@SiO₂-EDTA@Pt, 4750, 5464, 6794 and 10 789 for Fe₃O₄@SiO₂-DTPA@Pt. Furthermore, we catalyzed the hydrosilylation of methyldichlorosilane (180 mmol) and 1-hexene (110 mmol) using only 0.3 mg Fe₃O₄@SiO₂-EDTA@Pt (0.0052 μmol Pt) or 0.3 mg Fe₃O₄@SiO₂-DTPA@Pt (0.057 μmol Pt) in 12 h. Remarkably, the region-selectivity was still around 99% and TON increased to 662 733 for Fe₃O₄@SiO₂-EDTA@Pt, 579 947 for Fe₃O₄@SiO₂-DTPA@Pt, respectively.

Usually, the stability of a catalyst is an important characteristic before it can be used in an industrial application. Therefore, the reusability of Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt catalysts were evaluated through repetitively catalyzing the hydrosilylation of 1-hexene with methyldichlorosilane. A series of 12 consecutive hydrosilylation reactions were carried out under optimal reaction conditions. After each catalytic reaction, the two platinum catalysts were held in the original system using a magnet and catalyzed next hydrosilylation without any other process. The results were showed in Fig. 8. Data showed a satisfactory yield of 1-heptylmethyldichlorosilane (about 90% after 12 catalytic runs for both of Pt catalysts) which was apparent better than the Pt catalysts immobilized on non-magnetic SiO₂ supports (only 80% after 12 or 13 runs).^36,37 After 12 runs, the magnetism both of Pt catalysts were similar and the platinum amount was 155.78 mmol g⁻¹ for Fe₃O₄@SiO₂-EDTA@Pt and 177.62 mmol g⁻¹ for Fe₃O₄@SiO₂-DTPA@Pt. There were only 10.4% Pt loss in Fe₃O₄@SiO₂-EDTA@Pt and 7.9% Pt loss in Fe₃O₄@SiO₂-DTPA@Pt. The average loss of Pt was 0.87% for Fe₃O₄@SiO₂-EDTA@Pt and 0.66% for Fe₃O₄@SiO₂-DTPA@Pt in each reaction which was apparent better than homogeneous Pt catalyst (about 10 ppm) and Pt catalysts immobilized on non-magnetic support. The data were listed in Table S1.† In order to identify the active species of Pt catalysts, the state of Pt catalysts in high-resolution XPS after recycle experiments was tested. The data were listed in Fig. S3 and Table S2.† The total amount of Pt²⁺ and Pt⁰ was 71.1% (before recycle) and 76.1% (after recycle) for Fe₃O₄@SiO₂-EDTA@Pt. The total amount of Pt²⁺ and Pt⁰ was 73.6% (before recycle) and 79.6% (after recycle) for Fe₃O₄@SiO₂-DTPA@Pt. According to the data, there wasn't the apparent state change of the recycled Pt catalyst in both Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt. This also indicated both of catalysts, especially Fe₃O₄@SiO₂-DTPA@Pt, were very stable, which maybe result from strong coordination interaction between carboxyl group and the metallic Pt. The residual four carboxyl groups of DTPA in Fe₃O₄@SiO₂-DTPA maybe produce a stronger interaction with Pt than the residual three carboxyl groups of EDTA in Fe₃O₄@SiO₂-EDTA (Table 2).

Fig. 8 Catalyst recycle experiments for the hydrosilylation of 1-hexene with methyldichlorosilane.

Table 2 Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt-catalyzed tandem isomerization–hydrosilylation of trans- and cis-internal alkenes with methyldichlorosilane^a

Entry	Alkene	Yield^b (%)		Product	β/α
		a	b		a	b
a Conditions: alkenes, 10 mmol; methyldichlorosilane, 15 mmol; 8 h; a, Fe3O4@SiO2-EDTA@Pt as catalyst; b Fe3O4@SiO2-DTPA@Pt as catalyst.b yield were determined by GC using n-decane as an internal standard.
1		97	99		>20 : 1	>20 : 1
2		87	88		>20 : 1	>20 : 1
3		95	98		>20 : 1	>20 : 1
4		78	84		>20 : 1	>20 : 1
5		83	87		>20 : 1	>20 : 1

---

3.3.2 Catalyzing hydrosilylation of internal and terminal hexene. The unprecedented activity and selectivity of the two Pt nanoparticle catalysts prompted us to further exploit its potential applications in the isomerization–hydrosilylation tandem process. Five hexene, including 1-hexene, trans-2-hexene, cis-2-hexene, trans-3-hexene and cis-3-hexene were used as alkene. The hydrosilylation between methyldichlorosilane and hexene were catalyzed by Fe₃O₄@SiO₂-EDTA@Pt or Fe₃O₄@SiO₂-DTPA@Pt under the aforementioned optimal conditions. Products were monitored by GC and NMR (see Fig. S2 in ESI†). Data were summarized in Table 3. It was interesting that the main product was 1-hexylmethyldichlorosilane in hydrosilylation of all of hexene, which was typical isomerization–hydrosilylation and the β-adduct as a major product. The similar phenomena were observed in previous studies.^69,70 We bought the standard substance of 1-hexylmethyldichlorosilane which was compared with the synthesized 1-hexylmethyldichlorosilane, in GC chromatograms, all 1-hexylmethyldichlorosilane peek shown at around 8.45 min. For NMR: ¹H NMR (600 MHz, CDCl₃) δ: 1.50 (m, 2H, CH₃–(CH₂)₃–CH₂–CH₂–SiCl₂–CH₃), 1.38 (m, 2H, CH₃–(CH₂)₂–CH₂–(CH₂)₂–SiCl₂–CH₃), 1.20 (s, 4H, CH₃–(CH₂)₂–(CH₂)₃–SiCl₂–CH₃), 1.12 (m, 2H, CH₃–(CH₂)₄–CH₂–SiCl₂–CH₃), 0.88 (t, 3H, CH₃–(CH₂)₅–SiCl₂–CH₃), 0.77 (s, 3H, CH₃–(CH₂)₅–SiCl₂–CH₃).

Table 3 Hydrosilylation of alkenes with methyldichlorosilane catalyzed by Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt^a

Entry	Alkene	Structure	Time/h	Conv.^b (%)		TOF^c (10^3 h^−1)
				a	b	a	b
a Conditions: 15 mmol methyldichlorosilane, 10 mmol alkene; temperature: 60 °C; a, Fe3O4@SiO2-EDTA@Pt as catalyst (1.74 × 10^−3 mmol Pt); b, Fe3O4@SiO2-DTPA@Pt as catalyst (1.93 × 10^−3 mmol Pt).b The conversion was based on the amount of alkene consumed.c Initial TOFs were calculated by GC after 30 min of reaction time.
1	Cyclohexene		4	2	13	<0.1	<0.1
2	N,N-Dimethylacrylamide		4	96	98	8.9	8.2
3	1,2-Epoxy-4-vinylcyclohexane		4	99	98	9.5	9.3
4	Norbornylene		4	97	99	7.1	6.8
5	Methyl methacrylate		5	58	67	4.9	5.1
6	Butyl acrylate		5	57	87	4.1	3.8
7	Ethane-1,2-diyl bis(2-methylacrylate)		5	72	81	5.1	5.8
8	Crotonic acid ethyl ester		5	69	87	6.6	6.1
9	trans-2-Hexenyl acetate		5	68	97	5.6	4.9
10	trans-2-Hexen-1-yl phenylacetate		6	61	68	7.0	8.2
11	1,2-Epoxy-7-octene		6	98	99	8.7	9.1
12	Allyl glycidyl ether		6	84	98	5.1	4.4
13	5-Hexen-2-one		6	86	97	8.2	7.6
14	Allylchloride		4	58	64	6.2	6.1

---

The yield of 1-hexylmethyldichlorosilane changed in hydrosilylation of different hexane and methyldichlorosilane. When Fe₃O₄@SiO₂-EDTA@Pt was used as catalyst, the highest yield (97%) was from 1-hexene. The yield of 1-hexylmethyldichlorosilane decreased to 95% for cis-2-hexene and 83% for cis-3-hexene. When cis-2-hexene and cis-3-hexene were changed to trans-2-hexene and trans-3-hexene, the yield further decreased corresponding to 87% and 78%. The similar phenomenon was observed when Fe₃O₄@SiO₂-DTPA@Pt was used as catalyst. It indicated the hydrosilylation was harder for internal hexane than 1-hexene and trans-hexene than cis-hexane. The results also verified both of Pt catalysts could be used for the remote functionalization of alkenes, which has been shown to be a remarkable step in organic synthesis.^71,72

3.3.3 Catalyzing hydrosilylation of other alkenes. The catalytic activities of the two Pt nanoparticles were further evaluated through employing another 14 alkenes with different substituent, including nine terminal alkenes (N,N-dimethylacrylamide, 1,2-epoxy-4-vinylcyclohexane, methyl methacrylate, butyl acrylate, ethane-1,2-diyl bis(2-methylacrylate), 1,2-epoxy-7-octene, allyl glycidyl ether, 5-hexen-2-one and allylchloride), three trans- internal alkenes (crotonic acid ethyl ester, trans-2-hexenyl acetate, and trans-2-hexen-1-yl phenylacetate) and two ring type alkenes (cyclohexene and norbornylene). These compounds are important industry raw materials. For example, allyl glycidyl ether is an important substrate because the alkoxysilanes derived from this alkene have broad applications in coatings and as coupling agents for epoxy composites used as electronic chip encapsulation.⁷³ Catalytic results were summarized in Table 3. The multi-products were observed in GC chromatograms, different substrates peak at different times due to the diversity of their boiling point, for example, cyclohexene peek at around 6.86 min, 1,2-epoxy-4-vinylcyclohexane peek at around 9.55 min, and allyl glycidyl ether peek at around 8.18 min. Because of multi-products were observed, n-decane used as internal standard to get alkene conversion, which used to evaluate the versatility of the prepared catalysts. Various linear alkenes containing either electron-donating or electron-withdrawing groups such as amine (entry 2), ester (entries 5–10), epoxide (entries 3, 11, and 12), ketone (entry 13), alkyl chloride (entry 14) and ring type norbornylene (entry 4) except cyclohexene (entry 1) were well catalyzed by Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt. In catalyzing N,N-dimethylacrylamide, 1,2-epoxy-4-vinylcyclohexane, norbornylene, and 1,2-epoxy-7-octene, both Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt showed similar high activity (>96%). The catalytic activity of Fe₃O₄@SiO₂-DTPA@Pt was significantly higher than Fe₃O₄@SiO₂-EDTA@Pt in hydrosilylation of crotonic acid ethyl ester, trans-2-hexen-1-yl phenylacetate, trans-2-hexenyl acetate, methyl methacrylate, butyl acrylate, ethane-1,2-diyl bis(2-methylacrylate), allyl glycidyl ether, 5-hexen-2-one and allylchloride. Even Fe₃O₄@SiO₂-DTPA@Pt could catalyze the hydrosilylation of cyclohexene (13%), whereas a very low yield (2%) in Fe₃O₄@SiO₂-EDTA@Pt as catalyst. These results indicated Fe₃O₄@SiO₂-EDTA@Pt was more efficient.

4. Conclusions

Two novel magnetic Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt catalysts were successfully prepared. The saturation magnetization was 25.4 and 23.0 emu g⁻¹ for Fe₃O₄@SiO₂-EDTA@Pt and Fe₃O₄@SiO₂-DTPA@Pt, respectively. The Pt amount of the 30–50 nm nanoparticles was corresponding to 173.80 and 192.77 μmol g⁻¹, which the predominant valence was Pt²⁺ and Pt⁰. In catalyzing hydrosilylation of 1-hexene and methyldichlorosilane, the yield of 1-heptylmethyldichlorosilane was about 99% for both of catalysts and the TON were up to 662 733 for Fe₃O₄@SiO₂-EDTA@Pt and 579 947 for Fe₃O₄@SiO₂-DTPA@Pt, respectively. The two catalysts, especially Fe₃O₄@SiO₂-DTPA@Pt, showed good activity in hydrosilylation of 5 isomerous hexanes and 14 other alkene with different substituent. Fe₃O₄@SiO₂-DTPA@Pt even could catalyze the hydrosilylation of cyclohexene. After 12 catalytic runs for both of Pt catalysts, yield of 1-heptylmethyldichlorosilane was still up to 90% and the magnetism was similar. The average loss of Pt in each reaction was 0.87% for Fe₃O₄@SiO₂-EDTA@Pt and 0.66% for Fe₃O₄@SiO₂-DTPA@Pt. The unprecedented activity and selectivity of the two Pt nanoparticle catalysts are prospective and potential in catalyzing more reactions and replacing the current homogeneous Pt catalyst industry.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors would like to appreciate the financial supports from the National Natural Science foundation of China (No. 21605112).

References

1. R. M. Cornell and U. Schwertmann, The Iron Oxides: Structure, Properties, Reactions, Occurrence and Uses, VCH Publications, Weinheim, New York, 1996.
2. S. Xuan, Y. X. Wang, J. C. Yu and K. C. Leung, Langmuir, 2009, 25, 11835–11843.
3. B. Gaihre, M. S. Khil, D. R. Lee and H. Y. Kim, Int. J. Pharm., 2009, 365, 180–189.
4. N. Kohler, C. Sun, A. Fichtenholtz, J. Gunn, C. Fang and M. Zhang, Small, 2006, 2, 785–792.
5. H. Lee, M. K. Yu, S. Park, S. Moon, J. J. Min, Y. Y. Jeong, H. W. Kang and S. Jon, J. Am. Chem. Soc., 2007, 129, 12739–12745.
6. H. Liu, Z. Jia, S. Ji, Y. Zheng, M. Li and H. Yang, Catal. Today, 2011, 175, 293–298.
7. A. M. Donia, A. A. Atia and K. Z. Elwakeel, J. Hazard. Mater., 2008, 151, 372–379.
8. N. Nuryono, N. M. Rosiati, B. Rusdiarso, S. C. W. Sakti and S. Tanaka, SpringerPlus, 2014, 3, 515–526.
9. S. Hokkanen, E. Repo, S. Lou and M. Sillanpaa, Chem. Eng. J., 2015, 260, 886–894.
10. X. Sun, Q. Li, L. Yang and H. Liu, Particuology, 2016, 26, 79–86.
11. A. Khazaei, M. Khazaei and M. Nasrollahzadeh, Tetrahedron, 2017, 73, 5624–5633.
12. Y. Fang, Y. Chen, X. Li, X. Zhou, J. Li, W. Tang, J. Huang, J. Jin and J. Ma, J. Mol. Catal. A: Chem., 2014, 392, 16–21.
13. B. Panella, A. Vargas and A. Baiker, J. Catal., 2009, 261, 88–93.
14. R. L. Oliveira, P. K. Kiyohara and L. M. Rossi, Green Chem., 2010, 12, 144–149.
15. M. Amir, U. Kurtan and A. Baykal, Chin. J. Catal., 2015, 36, 1280–1286.
16. J. Sun, Z. Dong, X. Sun, P. Li, F. Zhang, W. Hu, H. Yang, H. Wang and R. Li, J. Mol. Catal. A: Chem., 2013, 367, 46–51.
17. J. J. Martíneza, E. X. Aguileraa, J. Cubillosa, H. Rojasa, A. Gómez-Cortésb and G. Díaz, Mol. Catal., 2019, 465, 54–60.
18. S. A. Theofanidis, V. V. Galvita, H. Poelman, R. Batchu, L. C. Buelens, C. Detavernier and G. B. Marin, J. Mol. Catal. B: Enzym., 2018, 239, 502–512.
19. Z. Mohamed, V. D. Dasireddy, S. Singh and H. B. Friedrich, Int. J. Hydrogen Energy, 2018, 43, 22291–22302.
20. B. Cheng, W. Liu and Z. Lu, J. Am. Chem. Soc., 2018, 140, 5014–5017.
21. B. Panella, A. Vargas and A. Baiker, J. Catal., 2009, 261, 88–93.
22. M. Walczak, K. Stefanowska, A. Franczyk, J. Walkowiak, A. Wawrzynczak and B. Marciniec, J. Catal., 2018, 367, 1–6.
23. T. Yang, Y. Huo, Y. Liu, Z. Rui and H. Ji, Appl. Catal., B, 2017, 200, 543–551.
24. H. A. Khan, P. Natarajan and K. Jung, Appl. Catal., B, 2018, 231, 151–160.
25. H. Zai, Y. Zhao, S. Chen, L. Ge and C. Chen, Nano Res., 2018, 61, 2544–2552.
26. Q. Li, S. Ji, M. Li and X. Duan, Sci. China Mater., 2018, 11, 1339–1344.
27. M. Mahmoudi, S. Sant, B. Wang, S. Laurent and T. Sen, Adv. Drug Delivery Rev., 2011, 63, 24–46.
28. R. T. Yang, Adsorbents: Fundamentals and Applications, John Wiley & Sons, USA, 2003.
29. A. Lu, E. L. Salabas and F. Schüth, Angew. Chem., Int. Ed., 2007, 46, 1222–1244.
30. G. Zeng, Y. Pang, Z. Zeng, L. Tang, Y. Zhang, Y. Liu, J. Zhang, X. Lei, Z. Li, Y. Xiong and G. Xie, Langmuir, 2012, 28, 468–473.
31. J. Ramo, M. Sillanpaa, V. Vickackaite, M. Orama and L. Niinisto, J. Pulp Pap. Sci., 2000, 26, 125–131.
32. J. Ramo and M. Sillanpaa, J. Cleaner Prod., 2001, 9, 191–195.
33. M. E. T. Sillanpaa and J. H. P. Ramo, Environ. Sci. Technol., 2001, 35, 1379–1384.
34. M. Sillanpaa, M. Orama, J. Ramo and A. Oikari, Sci. Total Environ., 2001, 267, 23–31.
35. K. Pirkanniemi, S. Metsarinne and M. Sillanpaa, J. Hazard. Mater., 2007, 147, 556–561.
36. F. Li and Y. Li, J. Mol. Catal. A: Chem., 2016, 420, 254–263.
37. D. Shao and Y. Li, RSC Adv., 2018, 8, 20379–20393.
38. Z. Lu, G. Wang, J. Zhuang and W. Yang, Colloids Surf., A, 2006, 278, 140–143.
39. Y. S. Kang, S. Risbud, J. F. Rabolt and P. Stroeve, Chem. Mater., 1996, 8, 2209–2211.
40. S. Zhang, H. Niu, Y. Zhang, J. Liu, Y. Shi, X. Zhang and Y. Cai, J. Chromatogr. A, 2012, 1238, 38–45.
41. Y. Deng, D. Qi, C. Deng, X. Zhang and D. Zhao, J. Am. Chem. Soc., 2008, 130, 28–29.
42. M. J. Jacinti, O. H. Santos, R. Landers, P. K. Kiyohara and L. M. Rossi, Appl. Catal., B, 2009, 90, 688–692.
43. Z. Ma, Y. Guan and H. Liu, J. Magn. Magn. Mater., 2006, 301, 469–477.
44. Q. Yuan, N. Li, Y. Chi, W. Geng, W. Yan, Y. Zhao, X. Li and B. Dong, J. Hazard. Mater., 2013, 254–255, 157–165.
45. X. Sun, Q. Li, L. Yang and H. Liu, Particuology, 2016, 26, 79–86.
46. M. J. Jacinto, P. K. Kiyohara, S. H. Masunaga, R. F. Jardim and L. M. Rossi, Appl. Catal., A, 2008, 338, 52–57.
47. M. Tülü and K. E. Geckeler, Polym. Int., 1999, 48, 909–914.
48. Y. Liu, M. Chen and Y. Hao, Chem. Eng. J., 2013, 218, 46–54.
49. Y. Shiraishi, G. Nishimura, A. T. Hirai and I. Komasawa, Ind. Eng. Chem. Res., 2002, 41, 5065–5070.
50. E. Repoa, J. K. Warcholc, T. A. Kurniawana and M. E. T. Sillanpää, Chem. Eng. J., 2010, 161, 73–82.
51. M. A. Hughes and E. Rosenberg, Sep. Sci. Technol., 2007, 42, 261–283.
52. G. Y. Li, K. L. Huang, Y. R. Jiang, D. L. Yang and P. Ding, Int. J. Biol. Macromol., 2008, 42, 405–412.
53. F. J. Arriagada and K. J. Osseo-Asare, J. Colloid Interface Sci., 1999, 211, 210–220.
54. J. Liu, Z. Sun, Y. Deng, Y. Zou, C. Li, X. Guo, L. Xiong, Y. Gao, F. Li and D. Zhao, Angew. Chem., Int. Ed., 2009, 48, 5875–5879.
55. H. Pan, X. Li, D. Zhang, Y. Guan and P. Wu, J. Mol. Catal. A: Chem., 2013, 277, 108–114.
56. S. Sayin and M. Yilmaz, J. Chem. Eng. Data, 2011, 56, 2020–2029.
57. Z. C. Li, H. T. Fan, Y. Zhang, M. X. Chen, Z. Y. Yu, X. Q. Cao and T. Sun, Chem. Eng. J., 2011, 171, 703–710.
58. H. Chen, C. Deng and X. Zhang, Angew. Chem., Int. Ed., 2010, 49, 607–611.
59. R. Qin, F. Li, M. Chen and W. Jiang, Appl. Surf. Sci., 2009, 256, 27–32.
60. A. I. Frenkel, C. W. Hills and R. G. Nuzzo, J. Phys. Chem. B, 2001, 105, 12689–12703.
61. G. Neri, C. Milone, S. Galvagno, A. P. J. Pijpers and J. Schwank, Appl. Catal., A, 2002, 227, 105–115.
62. R. Nie, D. Liang, L. Shen, J. Gao, P. Chen and Z. Hou, Appl. Catal., B, 2012, 127, 212–220.
63. Z. Yan, Z. Xu, Z. Yang, L. Yue and L. Huang, Appl. Surf. Sci., 2019, 467–468, 277–285.
64. J. Wang, S. Zheng, Y. Shao, J. Liu, Z. Xu and D. Zhu, J. Colloid Interface Sci., 2010, 349, 293–299.
65. F. Ge, M. M. Li, H. Ye and B. X. Zhao, J. Hazard. Mater., 2012, 211–212, 366–372.
66. C. J. Tan and Y. W. Tong, Anal. Chem., 2007, 79, 299–306.
67. D. Peng, Y. Zhang, X. Du, L. Zhang, X. Leng, M. D. Walter and Z. Huang, J. Am. Chem. Soc., 2013, 135, 19154–19166.
68. W. Hu, H. Xie, H. Yue, P. Prinsen and R. Luquee, Catal. Commun., 2017, 97, 51–55.
69. C. C. H. Atienza, T. Diao, K. J. Weller, S. A. Nye, K. M. Lewis, J. G. P. Delis, J. L. Boyer, A. K. Roy and P. J. Chirik, J. Am. Chem. Soc., 2014, 136, 12108–12118.
70. I. Buslov, F. Song and X. Hu, Angew. Chem., Int. Ed., 2016, 55, 12295–12299.
71. A. Vasseur, J. Bruffaerts and I. Marek, Nat. Chem., 2016, 8, 209–219.
72. R. Ciriminna, V. Pandarus, G. Gingras, F. Béland and M. Pagliaro, ACS Sustainable Chem. Eng., 2013, 1, 249–253.
73. E. A. Chernyshev, Z. V. Belyakova, L. K. Knyazeva and N. N. Khromykh, Russ. J. Gen. Chem., 2007, 77, 55–61.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ra00375d

---