Novel magnetic-separable and efficient Au/Fe–Al–O composite for the lactonization of 1,4-butanediol to γ-butyrolactone

Abstract

Nano-sized gold catalysts have attracted much interest during recent years. Here, we report a new material to improve the catalytic activity and recyclability of gold catalysts. Multi-component Fe–Al–O oxides prepared by impregnation (γ-Fe₂O₃/γ-Al₂O₃ and Al₂O₃/α-Fe₂O₃) and encapsulation (Fe₂O₃@Al₂O₃) methods were employed as carriers for gold catalysts. The Au/Fe–Al–O composites exhibited efficient catalytic activity and selectivity for the lactonization of 1,4-butanediol to γ-butyrolactone. They could also be easily recovered due to their super-paramagnetic property. XRD, TEM, XPS and H₂-TPR were used to characterize the composites. The influence of surface Fe/Al ratio on Au/(Fe₂O₃@Al₂O₃) composites was also studied by adjusting the Fe/Al molar ratio for comparison and the optimal Fe/Al ratio was found to be 1 : 4. The results confirmed that higher conversion and selectivity were obtained due to the bifunctional effect of Fe₂O₃ and Al₂O₃.

---

1. Introduction

Gold catalysts have been proven to be highly active in many reactions.^1–4 As previously reported, gold nanoparticles also exhibit excellent catalytic activity in the aerobic oxidations of alcohols,^5,6 like other supported metal catalysts.^7–10 Herein, the aerobic oxidation of 1,4-butanediol to γ-butyrolactone was studied not only because γ-butyrolactone is an important chemical that can be applied as a solvent, extraction agent or intermediate for the synthesis of many compounds,^11–13 but also because both its conversion and selectivity are related to the support effect and gold particle size effect according to our previous results.^14–17

Among many factors that affect the catalytic activity of gold catalysts,^18–23 it is of great significance to work on the supports because supports working in varied modes in gold catalysis can bring about great differences in gold particle size, geometry and oxidation state.^18,24 To improve the pore properties, acidic–basic properties and surface properties of mono-component supports, multi-component supports with two or more components are introduced.²⁵ It has been reported that the addition of a second element can provide reactive oxygen species for reactions, or stabilize the active gold species, increasing the catalytic activity²⁶ and stability^27–29 in many reactions, such as CO oxidation, oxidation of saturated hydrocarbons,^30–32 ammonia oxidation,³³ selective reduction of NO_x,³⁴ partial oxidation of methanol,²⁸ WGS reaction³⁵ and aerobic alcohol oxidation.³⁶ This kind of support material is widely used in gold catalysts, such as binary or ternary metal oxides^37–39 as Ce_1-xZr_xO₂ solid solution^40,41 and Al₂O₃,⁴² Fe₂O₃,³⁵ TiO₂^28,43,44 or CeO₂⁴⁵ modified with a main-group or transition metal oxide. G. J. Hutchings et al. recently reported that supported gold–palladium nanoparticles on carbon or TiO₂ are active for the oxidation of the primary carbon–hydrogen bonds in toluene and related molecules, as well as other bimetallic gold catalysts.^46,47 The influence of the gold nanoparticle size, nature of the support and influence of catalyst preparation procedure for the aerobic oxidation of alcohols were addressed to establish a detailed reaction mechanism that can help to design more active gold catalysts.⁴⁸ Recently, a study on the mechanism of ethanol and glycerol oxidation to acids over supported gold catalysts was reported by R. J. Davis et al.⁴⁹ M. Baumer et al. also reported an interesting preparation of nanoporous Au for selective gas-phase oxidative coupling of methanol.⁵⁰ More recent developments in the selective oxidation of organic compounds by gold catalysis have occurred during the last three years.^51,52

Although there are some reports about the gold-catalyzed oxidation of 1,4-butanediol to γ-butyrolactone, used both as the homogeneous one⁵ or the heterogeneous one,⁶ the multi-component support was not studied in detail. Herein, we used an environmental system for the oxidation of diols using air as the oxidizing agent. Lactones are a group of chemicals that can be applied as a solvent, extraction agent and intermediate of many biomedical products, fibers and pesticides, with wide application in agriculture, petroleum industry, pharmaceutics, resins and fibers. They can be synthesized by dehydrogenation or oxidation of the corresponding α,ω-diols and cycloketones. Current synthesis procedures involve fierce reaction conditions and oxidants, or the sacrifice of a series of cooxidants such as aldehydes, N-oxide, and so forth. Abundant wastes are also generated. Obviously, the process is not in accordance with the requirements of green chemistry and sustainable development.

In our previous studies, it was found that Au/Al₂O₃ shows high catalytic activity in the aerobic oxidation of 1,4-butanediol to γ-butyrolactone.¹⁷ Al₂O₃ is a common support for metal or metal oxide catalysts due to its stability, texture properties and unique acidic–basic properties.⁵³ The use of Al₂O₃ in the modification of TiO₂ has also been reported.^54,55 Yet, we also found that the selectivity of Au/Al₂O₃ increased in the initial stage and then dropped as the reaction went on. Also, none of the Au/Al₂O₃ catalysts gave a lactone yield higher than 90%. In the literature, a second component was added onto the surface of Al₂O₃ to prepare gold catalysts on multi-component supports containing Al₂O₃.

Aside from the enhancement of the activity of Au/Al₂O₃, our efforts have been focused on the development of a supported gold catalyst that can offer high efficiency combined with better recovery properties. A suitable solution to this problem is the utilization of the magnetic separation that enables the recovery of fine gold nanoparticles with higher catalytic activity.⁵⁶ For our Au/Fe–Al–O composites, separation of the substrate and recycling of the materials would be more convenient if ferromagnetic oxide is used as the support, since it can be simply collected by a permanent magnet due to its magnetic properties. Besides, iron oxide is a potential support material due to its reducibility, stability and the ability for electron transfer and Au/Fe₂O₃ showed high selectivity in our previous work.^14,15 Furthermore, considering its industrial application, the magnetic properties of our Au/Fe–Al–O composites can better solve the troubles of mass transfer in the three-phase system via the introduction of an alternating electromagnetic field. With an externally applied magnetic field, the catalysts can be well-dispersed and stabilized in the reaction system resistant to gravity so that they can make good contacts with the reagent even in a large scale system. Therefore, the advantages of the two metal composites were combined to develop a new material for a gold catalyst with high activity and selectivity for the oxidation of diols.

We addressed the different preparation methods to obtain fine materials. In the present work, Fe–Al–O supports with different morphologies and properties were synthesized. Impregnation and encapsulation methods were employed to prepare composites for gold catalysts. The properties of Au catalysts supported on different Fe–Al–O composites and their catalytic behaviors were explored and the influence of surface Fe/Al ratio on Fe₂O₃@Al₂O₃ catalysts was also investigated by adjusting the Fe/Al molar ratio.

2. Experimental section

2.1. Preparation of multi-component Fe–Al–O composites

The Fe–Al–O supports were prepared as described elsewhere.^57,58 The modified impregnation procedure is as follows: Al(NO₃)₃ and Fe(NO₃)₃ were respectively impregnated on the Fe₂O₃ or Al₂O₃ using ammonia aqueous solution as the precipitation agent, followed by calcination in air at 773 K. Samples impregnated from Fe(NO₃)₃ on Al₂O₃ are denoted as γ-Fe₂O₃/γ-Al₂O₃ and samples impregnated from Al(NO₃)₃ on Fe₂O₃ are denoted as Al₂O₃/α-Fe₂O₃.

The encapsulation procedure is as follows: ammonia aqueous solution was added into FeSO₄ aqueous solution. After the mixture was stirred at room temperature for 15 min, Al(NO₃)₃ was added dropwise and then Al(OH)₃ was precipitated on Fe(OH)₃. The as-prepared precipitates were collected by filtration, washed three times with deionized water and dried overnight at 373 K, followed by calcination in air. The as-obtained materials are denoted as mFe₂O₃@nAl₂O₃.

The formation process of the Au/Fe–Al–O composites is depicted in Scheme 1.

Scheme 1 A schematic illustration showing the possible formation mechanism of Au/Fe–Al–O composites.

2.2. Preparation of Au/Fe–Al–O composites

Deposition of gold onto the multi-component Fe–Al–O oxide support was carried out by the deposition–precipitation method (DP) using urea as the precipitation agent.^59,60 10 mL of HAuCl₄ aqueous solution (2.43 × 10⁻³ mol L⁻¹) was added into 50 mL of deionized water, followed by the addition of 2.92 g of urea. 0.6 g of Fe–Al–O oxides support was then added into the solution. The mixture was stirred for 2 h at 353 K. The as-prepared precipitates were collected by filtration, washed three times with deionized water and dried in air overnight at 373 K, followed by calcination in air in a muffle oven for 4 h at 573 K. The catalyst supported on the Fe–Al–O oxide support is denoted as Au/Fe–Al–O.

2.3. Characterization

The specific surface area of the samples was measured by nitrogen adsorption at 77 K (Micromeritics Tristar ASAP 3000) using the Brunauer–Emmett–Teller (BET) method. The gold loadings were determined by the inductively coupled plasma method (ICP, thermo E.IRIS). X-ray diffraction (XRD) patterns were recorded on a Bruker D8 Advance spectrometer with Cu-Kα radiation (λ = 0.154 nm), operated at 40 mA and 40 kV. Transmission electronic microscopy (TEM) measurement was performed on a JEOL JEM 2010 transmission electron microscope. The average size of the Au particles and their distributions were estimated by counting more than 300 Au particles. X-ray photoelectron spectra (XPS) were recorded under an ultra-high vacuum (<10⁻⁶ Pa) at a pass energy of 93.90 eV on a Perkin Elmer PHI 5000C ESCA system using Mg-Kα (1253.6 eV) anode. All binding energies were calibrated by using contaminant carbon (C1s = 284.6 eV). Temperature programmed reduction with hydrogen (H₂-TPR) of the samples was carried out in a full automatic XQ TP-5080 instrument (Tianjin Xianquan Co. Ltd.). About 50 mg of catalyst was packed into a reactor with quartz tubing and pretreated with high purity argon flow at 313 K for 2 h. Reduction was carried out under a mixture of hydrogen and argon (5% H₂) at a flow rate of 20 mL min⁻¹. The temperature was linearly raised at a ramping rate of 10 K min⁻¹ up to 973 K.

2.4. Activity measurements

The activity tests were carried out at 413 K in a stainless steel autoclave equipped with a magnetic stirrer. In a typical run of the reaction, 1.4 g of 1,4-butanediol was dissolved in 20 mL of tributyl phosphate (TBP), followed by addition of 0.190 g of Au/Fe–Al–O composite. Then, the autoclave was sealed and filled with 1.25 MPa air. The reaction was initiated by vigorous stirring. The rate of magnetic stirring was set at 800 rpm. The reaction mixture was sampled at regular time intervals and analyzed by gas chromatography (GC9560 (HuaAi Co. Ltd.) with a FID detector and HP5 capillary column) to determine the conversion and selectivity. Gas chromatography-mass spectrometry (GC 8000 TOF-MS voyager) and nuclear magnetic resonance (NMR, DMX 500) were utilized to determine the products. Scheme 2 shows the reaction equation.

Scheme 2 The reaction pathway of the lactonization of 1,4-butanediol to γ-butyrolactone.

3. Results and discussion

3.1. Characterization of the Fe–Al–O supports

The physico-chemical properties of the as-prepared Fe–Al–O support precursors are listed in Table 1. It was found that the BET specific surface area of the Fe₂O₃@Al₂O₃ support is higher than that of the other two supports prepared by the impregnation method, which are 153 and 163 m² g⁻¹ for γ-Fe₂O₃/γ-Al₂O₃ and Al₂O₃/α-Fe₂O₃, respectively. Iron oxide supports, including α-Fe₂O₃ and γ-Fe₂O₃ are in pure crystal phases, as evidenced by the XRD patterns of Fe–Al–O (Fig. 1). For Al₂O₃/α-Fe₂O₃, the XRD pattern shows structures assignable to hematite (α-Fe₂O₃). It can be inferred that the iron oxide is in the form of α-Fe₂O₃ after calcination at 773 K before impregnation and doesn't change after the impregnation. No crystalline structure of Al₂O₃ is observed, indicating that Al₂O₃ is mainly in the amorphous form. For γ-Fe₂O₃/γ-Al₂O₃ and Fe₂O₃@Al₂O₃, Fe₂O₃ is found to keep the crystal phase of γ-Fe₂O₃ even after calcination and weaker peaks of γ-Al₂O₃ (2θ = 45.3°, 66.8°) are found in the XRD patterns. The average particle size calculated from XRD patterns by Scherrer equation is also listed in Table 1. They are 20, 16 and 11 nm for Al₂O₃/α-Fe₂O₃, γ-Fe₂O₃/γ-Al₂O₃ and Fe₂O₃@Al₂O₃, respectively. The specific surface area of Fe₂O₃@Al₂O₃ is the highest (213 m² g⁻¹) due to it having the smallest particle size, which may provide a better surface for gold particles to disperse in the next steps.

Fig. 1 XRD patterns of Fe–Al–O: (a) amorphous Al₂O₃/α-Fe₂O₃, (b) γ-Fe₂O₃/γ-Al₂O₃ and (c) Fe₂O₃@Al₂O₃.

Table 1 Physico-chemical properties of Fe–Al–O prepared from different methods

Sample	Crystal phase	S BET (m^2 g^−1)	d (nm)^a
a Calculated by the Scherrer equation from the (311) peak (2θ = 35.6°) of maghemite or (104) peak (2θ = 33.2°) of hematite.
Al2O3/α-Fe2O3	α-Fe2O3	163	20
γ-Fe2O3/γ-Al2O3	γ-Fe2O3, γ-Al2O3	153	16
Fe2O3@Al2O3	γ-Fe2O3, γ-Al2O3	213	11

---

3.2. Characterization of Au/Fe–Al–O composites

Gold was deposited on Fe–Al–O supports by a homogeneous DP method. The physico-chemical properties of Au/Fe–Al–O composites are listed in Table 2. The support kept the same crystal phase and particle size after the deposition of gold, as observed from the XRD results (Fig. 2). XRD patterns display diffraction peaks indexed to α-Fe₂O₃ in Au/Al₂O₃/α-Fe₂O₃, while both γ-Fe₂O₃ and γ-Al₂O₃ in Au/γ-Fe₂O₃/γ-Al₂O₃ and Fe₂O₃@Al₂O₃, which coincides well with the H₂-TPR results. As presented in the H₂-TPR curves shown in Fig. 3, for Al₂O₃/α-Fe₂O₃ and γ-Fe₂O₃/γ-Al₂O₃, reduction peaks at 767 and 665 K for α-Fe₂O₃ and γ-Fe₂O₃ are due to the reduction of Fe(III) to Fe₃O₄. For Au/Fe₂O₃@Al₂O₃, the two reduction peaks show that Fe₂O₃ is in two morphologies, the bulk Fe₂O₃ and the surface Fe₂O₃ interacting with Al₂O₃.

Fig. 2 XRD patterns of Au/Fe–Al–O composites: (a) Au/Al₂O₃/α-Fe₂O₃, (b) Au/γ-Fe₂O₃/γ-Al₂O₃ and (c) Au/Fe₂O₃@Al₂O₃.

Fig. 3 H₂-TPR profiles of Au/Fe–Al–O composites: (a) Al₂O₃/α-Fe₂O₃, (b) Au/Al₂O₃/α-Fe₂O₃, (c) γ-Fe₂O₃/γ-Al₂O₃, (d) Au/γ-Fe₂O₃/γ-Al₂O₃, (e) Fe₂O₃@Al₂O₃ and (f) Au/ Fe₂O₃@Al₂O₃.

Table 2 Physico-chemical properties of Au/Fe–Al–O catalysts

Sample	Support crystal phase	S BET (m^2 g^−1)	d Au ^a (nm)
a Au particle size is determined by TEM. More than 300 particles were counted to give the average particle size.
Au/Al2O3/α-Fe2O3	α-Fe2O3	213	6.6
Au/γ-Fe2O3/γ-Al2O3	γ-Al2O3, γ-Fe2O3	142	4.6
Au/Fe2O3@Al2O3	γ-Al2O3, γ-Fe2O3	186	3.5

---

As shown in Fig. 2, the diffraction line at 2θ = 38.2 (Au(111)) indicates the formation of small metallic gold particles in the catalysts. Though gold diffraction peaks overlap with that of γ-Al₂O₃, the peaks of γ-Al₂O₃ are very weak and gold particle size can be judged from the XRD patterns. It shows that the particle size follows the order: Au/Al₂O₃/α-Fe₂O₃ > Au/γ-Fe₂O₃/γ-Al₂O₃ > Au/Fe₂O₃@Al₂O₃. More accurate and direct information on morphologies are provided by TEM images. Gold particle size and distributions differ among Au/Fe–Al–O composites, as shown in Fig. 4. For Au/Al₂O₃/α-Fe₂O₃ and Au/γ-Fe₂O₃/γ-Al₂O₃, the supports show a nanofiber morphology with Al₂O₃ and a nanoparticle morphology with Fe₂O₃, with an average diameter of 50 nm. For Fe₂O₃@Al₂O₃, Fe₂O₃ was encapsulated by Al₂O₃ and there were no isolated Fe₂O₃ particles observed, indicating that an interaction was formed between Al₂O₃ and Fe₂O₃. For Au/Al₂O₃/α-Fe₂O₃, gold particles are mainly dispersed on the Fe₂O₃ carrier, while those on the Al₂O₃ are larger in size with a range of particle size of 2–14 nm and average particle size of 6.6 nm. For Au/γ-Fe₂O₃/γ-Al₂O₃, gold particles are dispersed evenly on Al₂O₃ and Fe₂O₃ with much smaller average particle size of 4.6 nm. For Au/Fe₂O₃@Al₂O₃, gold particles are highly dispersed with the smallest average particle size of 3.5 nm.

Fig. 4 TEM images and Au particle size distributions of Au/Al₂O₃/α-Fe₂O₃ (a and b), Au/γ-Fe₂O₃/γ-Al₂O₃ (c and d) and Au/Fe₂O₃@Al₂O₃ (e and f).

Electronic properties of the Au/Fe–Al–O catalysts were characterized by XPS, as listed in Table 3. For Au/γ-Fe₂O₃/γ-Al₂O₃ and Au/Fe₂O₃@Al₂O₃, it can be found that the gold is in the metallic form (Au(0)), with only a single Au 4f_7/2 peak at a binding energy lower than 84.0 eV. But for Au/Al₂O₃/α-Fe₂O₃, more than half of the surface gold is in the oxidized state due to the stabilization of Au^δ+ by α-Fe₂O₃. The surface atomic ratio of Al, Fe and Au can also be obtained from the XPS results. Compared with the bulk atomic ratio of Fe and Al measured by ICP, for Au/Al₂O₃/α-Fe₂O₃ and Au/Fe₂O₃@Al₂O₃, the surface content of Al is higher than the bulk one, while that of Au/γ-Fe₂O₃/γ-Al₂O₃ is the lowest. For Au/Al₂O₃/α-Fe₂O₃, a higher surface content of Al is obtained since Al₂O₃ is impregnated on Fe₂O₃. For Au/Fe₂O₃@Al₂O₃, the surface content of Al is higher than that of Au/Al₂O₃/α-Fe₂O₃ as well as its bulk content, indicating that Fe₂O₃ is better encapsulated than in Au/Al₂O₃/α-Fe₂O₃. It is assumed that the rich hydroxyl groups on the surface of Al₂O₃ survived after the encapsulation method of Au/Fe₂O₃@Al₂O₃, which is beneficial to the generation of smaller gold particles according to our previous studies.¹⁷

Table 3 The binding energy and surface content of Au, Al, Fe in Au/Fe–Al–O composites

Sample	Au%surf	Au loading (%)^b	Fe:Alsurf^a	Fe:Albulk^b	Au 4f7/2 (eV)	Fe 2p3/2 (eV)	Al 2p3/2 (eV)
a Calculated from the XPS results. b Obtained from the ICP results.
Au/Al2O3/α-Fe2O3	1.8	6.2	1 : 10	1 : 4.1	83.3 (47%); 84.5 (53%)	711.9	74.8
Au/γ-Fe2O3/γ-Al2O3	5.5	5.9	1 : 3	1 : 2.2	83.3	710.9	74.0
Au/Fe2O3@Al2O3	4.1	6.0	1 : 18	1 : 3.4	83.8	712.0	74.7

---

3.3. Catalytic activity of Au/Fe–Al–O composites

The catalytic activity of the Au/Fe–Al–O composites and the simple mixture of Fe₂O₃ and Al₂O₃ in the selective aerobic oxidation of 1,4-butanediol to γ-butyrolactone are shown in Fig. 5 and Table 4. Au/Fe₂O₃@Al₂O₃ composite shows the highest activity. Two hour conversion of Au/Fe₂O₃@Al₂O₃ reaches 90% and its 4 h conversion is 99% with a selectivity of 87%. For Au/γ-Fe₂O₃/γ-Al₂O₃, 1,4-butanediol can be totally consumed in 8 h. Its catalytic activity is lower than Au/Al₂O₃+Au/Fe₂O₃ at the initial stage, but is much higher after 8 h reaction. The catalytic activity follows the order: Au/Fe₂O₃@Al₂O₃ > Au/γ-Fe₂O₃/γ-Al₂O₃ ≈ Au/Al₂O₃+Au/Fe₂O₃ > Au/Al₂O₃/α-Fe₂O₃. The selectivity of the Au/Fe–Al–O composites increases with the prolonging of the reaction time and keeps increasing after 8 h with no decline (Fig. 5(B)). Among them, Au/Fe₂O₃@Al₂O₃ catalyst shows the highest selectivity and reaches 99% at 8 h. The selectivity of Au/Al₂O₃/α-Fe₂O₃ and Au/Fe₂O₃+Au/Al₂O₃ increases slowly after 4 h reaction. The better catalytic performance of Au/Fe₂O₃@Al₂O₃ confirms that it is affected by the support properties and the Al₂O₃ and Fe₂O₃ are not a simple mixture, but interaction exists between them after the encapsulation process.

Fig. 5 Conversion (A) and selectivity (B) of oxidative dehydrogenation of 1,4-butanediol catalyzed by Au/Fe–Al–O catalysts. Reaction conditions: 0.190 g catalyst, 1.4 g 1,4-butanediol, 20 mL TBP, 1.25 MPa air, 413 K.

Table 4 Catalytic behavior of Au/Fe–Al–O composites in the oxidative dehydrogenation of 1,4-butanediol^a

Sample	Conversion (%)	Selectivity (%)	Yield (%)
a Reaction conditions: 0.190 g catalyst, 1.4 g 1,4-butanediol, 20 mL TBP, 1.25 MPa air, 413 K, 4 h.
Au/Al2O3/α-Fe2O3	88	76	67
Au/γ-Fe2O3/γ-Al2O3	93	88	82
Au Fe2O3@Al2O3	99	87	86
Au/Fe2O3+Au/Al2O3	93	84	78

---

3.4. The influence of Fe/Al ratio on Au/Fe₂O₃@Al₂O₃ composites

To gain an insight into the support effect and the interaction between Al₂O₃ and Fe₂O₃ in the Au/Fe₂O₃@Al₂O₃ composites, additional experiments were performed, probing the influence of Fe/Al ratio on the catalysts. The physico-chemical properties of Fe₂O₃@Al₂O₃ with different Fe/Al ratios are listed in Table 5. It was found that the specific surface area of Fe₂O₃@Al₂O₃ increases with a rise in Al content, from 131 to 236 m² g⁻¹. Except for Au/1Fe₂O₃@1Al₂O₃, the specific surface areas of other composites decrease little after the deposition of gold. The loading content of gold is lower than the theoretical content. It may be the weak interaction between Au and Al₂O₃ that makes Au difficult to deposit on the support, especially when the content of Al₂O₃ is high. XRD patterns (Fig. 6) display diffraction peaks indicative of γ-Fe₂O₃ and γ-Al₂O₃ in the support and no peaks of α-Fe₂O₃ are observed for all Au/Fe₂O₃@Al₂O₃ materials. With the increase in the content of Al, the diffraction peaks of γ-Al₂O₃ become stronger, while the peaks of γ-Fe₂O₃ become weaker.

Fig. 6 XRD patterns of Fe₂O₃@Al₂O₃ and Au/Fe₂O₃@Al₂O₃ composites with different Fe/Al ratios: (a) Fe₂O₃@Al₂O₃, (b) Au/1Fe₂O₃@1Al₂O₃, (c) 1Fe₂O₃@4Al₂O₃, (d) Au/1Fe₂O₃@4Al₂O₃, (e) 1Fe₂O₃@8Al₂O₃, (f) Au/1Fe₂O₃@8Al₂O₃, (g) 1Fe₂O₃@16Al₂O₃ and (h) Au/1Fe₂O₃@16Al₂O₃.

Table 5 Physico-chemical properties of Fe₂O₃@Al₂O₃ and Au/Fe₂O₃@Al₂O₃ with different Fe/Al ratios

Sample Fe:Al	Au loading (%)	S BETs ^a (m^2 g^−1)	S BETc ^b (m^2 g^−1)	d s ^c (nm)	d s ^d (nm)	d Au ^e (nm)
a BET surface area of the as-prepared Fe2O3@Al2O3 support. b BET surface area of the Au/Fe2O3@Al2O3 composites. c Calculated by the Scherrer equation from the (311) peak (2θ = 35.6°) of maghemite. d Particle size of the support in Au/Fe2O3@Al2O3 composites was calculated by the Scherrer equation from the (311) peak (2θ = 35.6°) of maghemite. e Au particle size was determined by TEM. More than 300 particles were counted to give the average particle size.
1 : 1	5.8	131	212	10	12	3.9
1 : 4	6.0	213	186	11	10	3.5
1 : 8	5.8	222	207	13	12	4.0
1 : 16	4.4	236	224	6.1	8.8	5.0

---

According to the XRD patterns, it can be inferred that the average particle size of gold follows the order: Au/1Fe₂O₃@16Al₂O₃ > Au/1Fe₂O₃@8Al₂O₃ > Au/1Fe₂O₃@1Al₂O₃ > Au/1Fe₂O₃@4Al₂O₃. More accurate data estimated by TEM are listed in Fig. 7. The TEM images show that the supports are composed of nanofibers with no nanoparticles observed in the Au/1Fe₂O₃@4Al₂O₃, Au/1Fe₂O₃@8Al₂O₃ and Au/1Fe₂O₃@16Al₂O₃ composites. For Au/1Fe₂O₃@1Al₂O₃, it is assumed from the nanoparticles shown in the TEM images that Fe₂O₃ was not well encapsulated by Al₂O₃ due to low Al content. The content of Al is exact to cover all the Fe₂O₃ particles in the Au/1Fe₂O₃@4Al₂O₃. It is found that gold particles are highly dispersed on the support with a small particle size. For Au/1Fe₂O₃@1Al₂O₃, Au/1Fe₂O₃@4Al₂O₃ and Au/1Fe₂O₃@8Al₂O₃, gold particles are dispersed evenly on Al₂O₃ and Fe₂O₃ with a range of particle size of 1–9 nm, which are 3.9, 3.5 and 4 nm, respectively. For Au/1Fe₂O₃@16Al₂O₃, the average particle size of gold is much larger at 5.0 nm.

Fig. 7 TEM images and gold particle size distributions of the Au/Fe₂O₃@Al₂O₃ composites with different Fe/Al ratios: 1 : 1 (a and b), 1 : 4 (c and d), 1 : 8 (e and f) and 1 : 16 (g and h).

The electronic properties and surface content of the Au/Fe₂O₃@Al₂O₃ with different Fe/Al ratios were characterized by XPS and ICP, as listed in Table 6. It can be found that gold is in the metallic form (Au(0)) with only a single Au 4f_7/2 peak at binding energy of 83.5–83.9 eV. According to the XPS and ICP results, the surface content of Al is 3–4 times higher than the bulk one for Au/Fe₂O₃@Al₂O₃. The surface content of Al has a maximum value at a certain Fe/Al ratio.

Table 6 The binding energy and surface content of Au, Al, Fe in Au/Fe₂O₃@Al₂O₃ with different Fe/Al ratios

Sample Fe:Al	Au loading (%)	S BETs ^a (m^2 g^−1)	S BETc ^b (m^2 g^−1)	d s ^c (nm)	d s ^d (nm)	d Au ^e (nm)
a Calculated from the XPS results. b Obtained from the ICP results.
1 : 1	5.8	131	212	10	12	3.9
1 : 4	6.0	213	186	11	10	3.5
1 : 8	5.8	222	207	13	12	4.0
1 : 16	4.4	236	224	6.1	8.8	5.0

---

The reactivity of the Au/Fe₂O₃@Al₂O₃ catalysts with different Fe/Al ratios is shown in Fig. 8 and Table 7. All of the Au/Fe₂O₃@Al₂O₃ catalysts show higher activity than Au/Fe₂O₃ and Au/FeO_x–NF, as reported previously.^14,15 The catalytic activity order is as follows: Au/1Fe₂O₃@4Al₂O₃ > Au/1Fe₂O₃@8Al₂O₃ > Au/1Fe₂O₃@16Al₂O₃ > Au/1Fe₂O₃@1Al₂O₃. Four hour conversions of Au/1Fe₂O₃@4Al₂O₃ and Au/1Fe₂O₃@8Al₂O₃ reach 99% and their yields reach 80%, which are very close to the value of Au/Al₂O₃. Compared with Au/Al₂O₃, Au/Fe₂O₃@Al₂O₃ catalysts show higher stability during the reaction. Their selectivity keeps rising with the prolonging of reaction time, while that of Au/Al₂O₃ drops immediately after 4h. The Au/1Fe₂O₃@4Al₂O₃ obtained a conversion of 99% and selectivity of 93% in 8 h, with a shortened reaction time and higher selectivity than the maximum one of Au/Al₂O₃. Unlike the other Au/Fe₂O₃@Al₂O₃ catalysts, the selectivity of Au/1Fe₂O₃@16Al₂O₃ does not increase with the prolonging of the reaction time (Fig. 8 (B)); instead, it increases in the initial 4 h and then drops a little as the reaction goes on, like Au/Al₂O₃. The decline in the selectivity after 4 h is assumed to be because the low content of Fe leads to the stronger support basicity, which is not beneficial to the reaction because it helps the products’ hydrolysis, thus decreasing the selectivity. Maximum selectivity and yield are obtained at 4 h reaction time.

Fig. 8 Conversion (A) and selectivity (B) of the dehydrogenation of 1,4-butanediol catalyzed by the Au/Fe₂O₃@Al₂O₃ composites with different Fe/Al ratios. Reaction conditions: 0.190 g catalyst, 1.4 g 1,4-butanediol, 20 mL TBP, 1.25 MPa air, 413 K.

Table 7 Catalytic behavior of Au/Fe₂O₃@Al₂O₃ with different Fe/Al ratios in the oxidative dehydrogenation of 1,4-butanediol^a

Sample	Conversion (%)	Selectivity (%)	Yield (%)
a Reaction conditions: 0.190 g of catalyst, 1.4 g of 1,4-butanediol, 20 mL of TBP, 1.25 MPa air, 413 K, 4 h.
Au/Fe2O3	83	76	62
Au/1Fe2O3@1Al2O3	89	70	62
Au/1Fe2O3@4Al2O3	99	87	86
Au/1Fe2O3@8Al2O3	99	82	81
Au/1Fe2O3@16Al2O3	95	85	81
Au/Al2O3	100	88	88

---

The above studies on Fe/Al ratios indicate the influence of the interaction between Fe₂O₃ and Al₂O₃. For Au/1Fe₂O₃@1Al₂O₃, the content of alumina is not enough to encapsulate iron oxide totally, corresponding to its lower activity which is close to Au/ γ-Fe₂O₃/γ-Al₂O₃ and Au/(Fe₂O₃+Al₂O₃) catalysts. Alumina can encapsulate iron oxide thoroughly in Au/1Fe₂O₃@4Al₂O₃ and Au/1Fe₂O₃@8Al₂O₃ composites with smaller gold particles, in correlation to the high activity in the reaction. For Au/1Fe₂O₃@16Al₂O₃, isolated Al₂O₃ particles on the surface may result from the low content of iron oxide and thus the properties of the support are close to Au/Al₂O₃. However, its gold particle size is larger than that of Au/Al₂O₃ and thus it is less reactive than Au/Al₂O₃. As shown in the above results, the Au/1Fe₂O₃@4Al₂O₃ composite, which was successfully obtained with optimized Fe/Al ratio, showed the highest catalytic activity and selectivity of all the materials synthesized in this work. Further work on the improvement of selectivity to lactones is under way.

3.5. The recyclable use of Au/Fe₂O₃@Al₂O₃ composites

Fig. 9 shows the photos of the magnetically induced separation of Au/Fe₂O₃@Al₂O₃ by exploiting their magnetic properties. A total amount of 0.01 g of sample was dispersed in 5 mL reaction mixture and the suspension was sonicated for 5 min. The resultant suspension appeared to be dark brown in color (Fig. 9a). When placing the vial near an external magnet for 1 min, the progressive separation of catalytic NPs was observed, while the composite was attracted toward the wall of the vial, which was closer to the magnet (Fig. 9b). With the time duration was extended to 10 min, the suspension became totally clear (Fig. 9c). Besides that, the composite powder can be re-dispersed again simply by shaking after removing the magnet (Fig. 9d).

Fig. 9 Pictures of the progressive separation of Au/Fe₂O₃@Al₂O₃ from the suspension (a) without magnet, (b) upon application of a magnet for 1 min, (c) upon application of a magnet for 10 min and (d) re-dispersion by shaking after removing the magnet.

This shows the potential application of our Au/Fe₂O₃@Al₂O₃ composites in the industry. On one hand, the magnetic properties of the Au/Fe₂O₃@Al₂O₃ composites can make the catalyst easily well-dispersed in the three-phase reaction system via the introduction of an alternating electromagnetic field. With the externally applied magnetic field, the catalysts can be stably suspended in the reaction system resistant to gravity so that they would make good contacts with the reagent even in a large scale system. On the other hand, after the reaction, the catalysts could also be easily recovered by magnetic force, by means of the externally applied magnetic field. Hence, this kind of magnetic catalytic nanoparticle has the potential to be easily recovered after liquid phase reaction and greatly facilitate the practical running of an industrial plant. In addition, we have detected the metal content in the reaction mixture after the magnet separation by means of an ICP method and found that the remaining metal concentration is below the detection limit.

The Au/Fe₂O₃@Al₂O₃ composite could be reused for 4 successive reactions without significant loss of the catalytic activity (Fig. 10). In the XPS spectra, phosphorus is found on the surface of the recycled materials, as well as gold, aluminium and iron. The presence of phosphorus may be due to the adsorption of solvent, TBP (C₁₂H₂₇O₄P), on the composites after reaction. Thus, it is assumed that the remaining TBP covers the active sites of the composites, resulting in the decrease in the reactivity. However, further studies on the regeneration will proceed in the future.

Fig. 10 Recyclability of the oxidative dehydrogenation of 1,4-butanediol catalyzed by Au/Fe₂O₃@Al₂O₃ composites with different Fe/Al ratios. Reaction conditions: catalyst:1,4-butanediol = 1 : 200, 20 mL TBP, 1.25 MPa air, 413 K.

4. Conclusions

In this work, we synthesized a new support to improve the catalytic activity and recovery properties of gold catalysts. Multi-component Fe–Al–O composites prepared by encapsulation (Fe₂O₃@Al₂O₃) and impregnation methods (γ-Fe₂O₃/γ-Al₂O₃ and Al₂O₃/α-Fe₂O₃) are employed as supports for gold catalysts. They were systematically studied. The Au/Fe–Al–O composites exhibited efficient catalytic activity and selectivity for the lactonization of 1,4-butanediol to γ-butyrolactone. The materials could also be easily recovered due to its super-paramagnetic property.

Compared with the composites synthesized by the impregnation method, Au/Fe₂O₃@Al₂O₃ showed a higher conversion and selectivity in the oxidative dehydrogenation of 1,4-butanediol and its catalytic activity was better than that of the mixture of Au/Fe₂O₃ and Au/Al₂O₃, very close to that of the most reactive Au/Al₂O₃ catalyst. Gold can be highly dispersed on the Fe₂O₃@Al₂O₃ support surface with an average diameter of 3.5 nm and gold–support interaction was observed because its surface was enriched by hydroxyl groups like Al₂O₃ and basicity could be adjusted by Fe₂O₃.

By adjusting the Fe/Al molar ratio in Au/Fe₂O₃@Al₂O₃ composites, supports with different surface Fe/Al ratios were obtained. A lower amount of Al₂O₃ was not enough to encapsulate all of the Fe₂O₃ particles and the optimized Fe/Al ratio was found to be 1 : 4. In addition, a further increase in Al content would lead to the weakening of the Fe–Al interaction and stronger support basicity, which was disadvantageous to the reaction.

Acknowledgements

Financially supported by the Major State Basic Resource Development Program (Grant No. 2012CB224804), NNSFC (Project 21173052, 20973042), the Research Fund for the Doctoral Program of Higher Education (20090071110011) and the Natural Science Foundation of Shanghai Science & Technology Committee (08DZ2270500).

References

1. A. Abad, P. Concepcion, A. Corma and H. Garcia, Angew. Chem., Int. Ed., 2005, 44, 4066–4069.
2. D. Andreeva, V. Idakiev, T. Tabakova, A. Andreev and R. Giovanoli, Appl. Catal., A, 1996, 134, 275–283.
3. M. Haruta, N. Yamada, T. Kobayashi and S. Iijima, J. Catal., 1989, 115, 301–309.
4. M. D. Hughes, Y. J. Xu, P. Jenkins, P. McMorn, P. Landon, D. I. Enache, A. F. Carley, G. A. Attard, G. J. Hutchings, F. King, E. H. Stitt, P. Johnston, K. Griffin and C. J. Kiely, Nature, 2005, 437, 1132–1135.
5. W. B. Hou, N. A. Dehm and R. W. J. Scott, J. Catal., 2008, 253, 22–27.
6. T. Mitsudome, A. Noujima, T. Mizugaki, K. Jitsukawa and K. Kaneda, Green Chem., 2009, 11, 793–797.
7. H. B. Ji, T. Mizugaki, K. Ebitani and K. Kaneda, Tetrahedron Lett., 2002, 43, 7179–7183.
8. M. S. Kwon, N. Kim, C. M. Park, J. S. Lee, K. Y. Kang and J. Park, Org. Lett., 2005, 7, 1077–1079.
9. T. Nishimura, N. Kakiuchi, M. Inoue and S. Uemura, Chem. Commun., 2000, 1245–1246.
10. N. Kakiuchi, T. Takahiro Nishimura, M. Masashi Inoue and S. Sakae Uemura, Bull. Chem. Soc. Jpn., 2001, 74, 165–172.
11. A. C. Albertsson and I. K. Varma, Biomacromolecules, 2003, 4, 1466–1486.
12. S. Kobayashi, Macromol. Symp., 2006, 240, 178–185.
13. H. Lee, F. Q. Zeng, M. Dunne and C. Allen, Biomacromolecules, 2005, 6, 3119–3128.
14. J. Huang, W. L. Dai and K. N. Fan, J. Phys. Chem. C, 2008, 112, 16110–16117.
15. J. Huang, W. L. Dai and K. N. Fan, J. Catal., 2009, 266, 228–235.
16. J. Huang, W. L. Dai, H. X. Li and K. Fan, J. Catal., 2007, 252, 69–76.
17. J. Huang, Y. Wang, J. M. Zheng, W. L. Dai and K. N. Fan, Appl. Catal., B, 2011, 103, 343–350.
18. M. Comotti, W. C. Li, B. Spliethoff and F. Schuth, J. Am. Chem. Soc., 2006, 128, 917–924.
19. A. C. Gluhoi, N. Bogdanchikova and B. E. Nieuwenhuys, J. Catal., 2005, 229, 154–162.
20. M. M. Schubert, S. Hackenberg, A. C. van Veen, M. Muhler, V. Plzak and R. J. Behm, J. Catal., 2001, 197, 113–122.
21. W. F. Yan, B. Chen, S. M. Mahurin, V. Schwartz, D. R. Mullins, A. R. Lupini, S. J. Pennycook, S. Dai and S. H. Overbury, J. Phys. Chem. B, 2005, 109, 10676–10685.
22. J. Q. Lu, X. M. Zhang, J. J. Bravo-Suarez, K. K. Bando, T. Fujitani and S. T. Oyama, J. Catal., 2007, 250, 350–359.
23. . J. Guzman and B. C. Gates, J. Am. Chem. Soc., 2004, 126, 2672–2673.
24. B. E. Solsona, T. Garcia, C. Jones, S. H. Taylor, A. F. Carley and G. J. Hutchings, Appl. Catal., A, 2006, 312, 67–76.
25. M. J. Lippits, A. C. Gluhoi and B. E. Nieuwenhuys, Top. Catal., 2007, 44, 159–165.
26. R. J. H. Grisel, J. J. Slyconish and B. E. Nieuwenhuys, Top. Catal., 2001, 16, 425–431.
27. M. A. Centeno, K. Hadjiivanov, T. Venkov, H. Klimev and J. A. Odriozola, J. Mol. Catal. A: Chem., 2006, 252, 142–149.
28. F. W. Chang, H. Y. Yu, L. S. Roselin, H. C. Yang and T. C. Ou, Appl. Catal., A, 2006, 302, 157–167.
29. K. Qian, W. X. Huang, Z. Q. Jiang and H. X. Sun, J. Catal., 2007, 248, 137–141.
30. M. A. Centeno, M. Paulis, M. Montes and J. A. Odriozola, Appl. Catal., A, 2002, 234, 65–78.
31. A. C. Gluhoi, N. Bogdanchikova and B. E. Nieuwenhuys, J. Catal., 2005, 232, 96–101.
32. A. C. Gluhoi and B. E. Nieutwenhuys, Catal. Today, 2007, 119, 305–310.
33. S. D. Lin, A. C. Gluhoi and B. E. Nieuwenhuys, Catal. Today, 2004, 90, 3–14.
34. D. Niakolas, C. Andronikou, C. Papadopoulou and H. Matralis, Catal. Today, 2006, 112, 184–187.
35. J. M. Hua, Q. Zheng, Y. H. Zheng, K. M. Wei and X. Y. Lin, Catal. Lett., 2005, 102, 99–108.
36. P. Haider and A. Baiker, J. Catal., 2007, 248, 175–187.
37. C. Gennequin, M. Lamallem, R. Cousin, S. Siffert, V. Idakiev, T. Tabakova, A. Aboukais and B. L. Su, J. Mater. Sci., 2009, 44, 6654–6662.
38. P. Haider, J. D. Grunwaldt, R. Seidel and A. Baiker, J. Catal., 2007, 250, 313–323.
39. V. Idakiev, T. Tabakova, K. Tenchev, Z. Y. Yuan, T. Z. Ren, A. Vantomme and B. L. Su, J. Mater. Sci., 2009, 44, 6637–6643.
40. I. Dobrosz-Gomez, I. Kocemba and J. M. Rynkowski, Appl. Catal., B, 2008, 83, 240–255.
41. I. Dobrosz-Gomez, I. Kocemba and J. M. Rynkowski, Catal. Lett., 2009, 128, 297–306.
42. R. J. H. Grisel and B. E. Neiuwenhuys, Catal. Today, 2001, 64, 69–81.
43. Z. Ma, S. H. Overbury and S. Dai, J. Mol. Catal. A: Chem., 2007, 273, 186–197.
44. V. Rodriguez-Gonzalez, R. Zanella, L. A. Calzada and R. Gomez, J. Phys. Chem. C, 2009, 113, 8911–8917.
45. G. Avgouropoulos, M. Manzoli, F. Boccuzzi, T. Tabakova, J. Papavasiliou, T. Ioannides and V. Idakiev, J. Catal., 2008, 256, 237–247.
46. L. Kesavan, R. Tiruvalam, M. H. A. Rahim, M. I. bin Saiman, D. I. Enache, R. L. Jenkins, N. Dimitratos, J. A. Lopez-Sanchez, S. H. Taylor, D. W. Knight, C. J. Kiely and G. J. Hutchings, Science, 2011, 331, 195–199.
47. N. Dimitratos, J. A. Lopez-Sanchez and G. J. Hutchings, Chem. Sci., 2012, 3, 20–44.
48. A. Abad, A. Corma and H. García, Chem.–Eur. J., 2008, 14, 212–222.
49. B. N. Zope, D. D. Hibbitts, M. Neurock and R. J. Davis, Science, 2010, 330, 74–78.
50. A. Wittstock1, V. Zielasek1, J. Biener, C. M. Friend and M. Bäumer, Science, 2010, 327, 319–322.
51. C. D. Pina, E. Falletta and M. Rossi, Chem. Soc. Rev., 2012, 41, 350–369.
52. C. P. Vinod, K. Wilson and A. F. Lee, J. Chem. Technol. Biotechnol., 2011, 86, 161–171.
53. Y. F. Han, Z. Y. Zhong, K. Ramesh, F. X. Chen and L. W. Chen, J. Phys. Chem. C, 2007, 111, 3163–3170.
54. C. M. Wang, K. N. Fan and Z. P. Liu, J. Phys. Chem. C, 2007, 111, 13539–13546.
55. W. F. Yan, S. M. Mahurin, Z. W. Pan, S. H. Overbury and S. Dai, J. Am. Chem. Soc., 2005, 127, 10480–10481.
56. V. Polshettiwar, R. Luque, A. Fihri, H. Zhu, M. Bouhrara and J. Basset, Chem. Rev., 2011, 111, 3036–3075.
57. M. A. Karakassides, D. Gournis, A. B. Bourlinos, P. N. Trikalitis and T. Bakas, J. Mater. Chem., 2003, 13, 871–876.
58. K. Mori, S. Kanai, T. Hara, T. Mizugaki, K. Ebitani, K. Jitsukawa and K. Kaneda, Chem. Mater., 2007, 19, 1249–1256.
59. R. Zanella, S. Giorgio, C.-H. Shin, C. R. Henry and C. Louis, J. Catal., 2004, 222, 357–367.
60. J. Radnik, L. Wilde, M. Schneider, M. M. Pohl and D. Herein, J. Phys. Chem. B, 2006, 110, 23688–23693.

---