Fabrication and Catalytic Activity of Thermally Stable Gold Nanoparticles on Ultrastable Y (USY) Zeolites

Abstract

Au was deposited on ultrastable Y (USY) zeolites using an ion-exchange method. Up to 5.5 wt% Au was introduced into the NH₄-form of USY zeolites. In contrast, deposition of Au hardly took place on the H- and Na-forms of Y-type zeolites, NH₄-forms of mordenite, and ZSM-5. Treatment of the Au-loaded USY zeolite in a H₂ atmosphere, afforded Au⁰ nanoparticles. These particles were thermally stable even at 973 K, where their mean particle diameter was 3.7 nm. In contrast, highly aggregated Au particles were observed after thermal treatment at temperatures lower than 523 K, followed by storage in air for a month. The resulting particle sizes were in good correlation with the IR band intensity of the adsorbed CO and the catalytic activity of Au in the aerobic oxidation of benzyl alcohol. The Au nanoparticles showed highest activity when the Au/USY zeolite was thermally treated at 673–973 K. A negligible deactivation was observed after repeating the reaction at least 12 times. In the case of Au/TiO₂ catalyst prepared by the deposition-precipitation method, the highest activity was observed at 573 K, which was lower than the temperature used for the Au/USY zeolites. This study demonstrated the potential use of the NH₄-form of USY zeolites for supporting Au.

1. Introduction

Much attention has been paid to Au-loaded catalysts since the pioneering work done by Haruta and co-researchers [1,2]. It has been reported that the catalytic activity of Au nanoparticles is sensitive to the preparation method and the type of support because it depends strongly on the size and shape of the Au nanoparticles [3]. In particular, high catalytic activity has been reported for the Au particles smaller than several nanometers in size, not only in the research field of heterogeneous, but also in the homogeneous catalysts [4,5,6]. Earlier reports dealing with heterogeneous Au catalysts focused on the use of reducible metal oxides, such as TiO₂ as a support for Au. To date, several methods, including deposition-precipitation [7], chemical vapor deposition [8], and cation adsorption [9], have been applied for the preparation of Au/TiO₂ catalysts. Among these methods, deposition-precipitation has been the most frequently employed method for Au loading, and it involves the use of tetrachloroauric (III) acid as a precursor of Au. In this method, the pH of tetrachloroauric (III) acid solution was adjusted by an aqueous solution of ammonia, NaOH, and urea in the presence of supports such as TiO₂ [10,11]. On the other hand, the volume of research dealing with Au-loaded acidic supports is smaller than that on basic supports, possibly because the deposition of AuCl₃ on acidic supports was difficult in an aqueous alkaline solution that was otherwise employed for Au loading on TiO₂. Despite this challenge, several groups have reported the use of acidic zeolites as supports for Au. For instance, Au loaded on Y-type and HZSM-5 zeolites was used in the direct synthesis of hydrogen peroxide [12]. The Au(en)₂Cl₃ complex was supported on Y-type zeolites to obtain electron-deficient Au particles [13]. Au nanoparticles, accessible only through zeolite micropores, were reported to be active in the size-selective aerobic aldehyde oxidation [14]. Zeolites are promising supports for Au because of the presence of ion-exchange sites such as NH₄⁺. It was expected that Au could be introduced into the NH₄⁺-zeolite through the formation of NH₄Cl, providing a new and simple method for introducing Au into the zeolite by mixing a solution of HAuCl₄ and NH₄⁺-zeolite without any special care. In other words, Cl⁻ of HAuCl₄ was removed through the reaction with NH₄⁺, unlike in the case of impregnated samples wherein Cl⁻ remained on the support [15]. Other advantages of zeolite include high thermal stability, presence of large surface area, and micropores that may hinder the sintering of Au. Also, coalescence of Au may be thwarted by the wall structure of zeolite at high temperatures.

The purpose of this study is to establish a new method for the preparation of Au nanoclusters on zeolite supports and demonstrate that Au° nanoparticles having a narrow size distribution were obtained after calcination at >673 K. Here, ultrastable Y (USY) zeolites were predominantly employed as the supports for Au. The USY zeolites were prepared by the steam treatment of NH₄⁺-Y, followed by reaction with an ammonium nitrate solution. The resulting ammonium-treated USY zeolites had strong Brønsted acid sites [16]. First, we developed a new method for loading Au onto NH₄⁺-USY zeolites using an aqueous solution of HAuCl₄. Then, the as-prepared Au/USY zeolites were thermally treated in a H₂ atmosphere at 353–1073 K. The treated samples were characterized by X-ray diffraction (XRD), transmission electron microscopy (TEM), and infrared (IR) spectroscopy. Partial oxidation of benzyl alcohol in the liquid phase was employed as a test reaction on the supported Au. This is because the oxidation of benzyl alcohol has received substantial attention as a probe molecule to test the selectivity and efficiency of novel metallic Au catalysts [17]. The catalytic activity and selectivity of Au nanoparticles were sensitive to the particle size and electronic state of Au [18]. The catalytic activity of the thermally treated Au/USY zeolite was correlated with the Au dispersion, as measured by XRD and TEM. The performance was compared with that of Au/TiO₂ catalyst prepared by the conventional deposition-precipitation method.

2. Results and Discussion

2.1. Loading of Au on Y-Type Zeolites

Table 1 lists the values of Au loading as measured by inductively coupled plasma (ICP) after digestion of zeolite-supported Au with aqua regia. Most of the Au present in the mother liquor was introduced into the NH₄⁺-Y, CaNH₄⁺-Y, and NH₄⁺-USY zeolites. Deposition of Au partially took place on β-type zeolite (1.8 wt%). In contrast, the amounts of Au on the NH₄⁺-ZSM-5, NH₄⁺-mordenite, and Na⁺-Y zeolites were almost negligible. Probably, a supercage—a relatively large space with a diameter of 1.3 nm—was necessary to accommodate HAuCl₄ in Y-type zeolites.

Table 1. Si/Al₂ ratio and Au loading of Au/zeolites.

Support	Si/Al2	Loading/wt%	Support	Si/Al2	Loading/wt%
NH4-ZSM-5	90	0.1	H-USY	7.7	0.1
NH4-MOR	15	0.1	NH4-Y	5.5	3.1
NH4-beta	20	1.8	NH4-CaY	7.7	3.3
Na-Y	5.5	0.1	NH4-USY	7.7	3.1

Table 1. Si/Al₂ ratio and Au loading of Au/zeolites.

Support	Si/Al2	Loading/wt%	Support	Si/Al2	Loading/wt%
NH4-ZSM-5	90	0.1	H-USY	7.7	0.1
NH4-MOR	15	0.1	NH4-Y	5.5	3.1
NH4-beta	20	1.8	NH4-CaY	7.7	3.3
Na-Y	5.5	0.1	NH4-USY	7.7	3.1

Figure 1 (▲) shows Au loading on USY zeolites that were thermally treated in a N₂ stream at different temperatures prior to Au loading. Approximately 3 wt% of Au was loaded onto the USY zeolites that had been treated at <523 K. The amount of Au loading decreased on increasing the temperature to 573 K. Only 0.1 wt% of Au was loaded onto the USY zeolite at 673–773 K. The amount of NH₄⁺ was measured by the temperature programmed desorption of NH₃ at each temperature and plotted in Figure 1 (〇). The amount of NH₄⁺ in the USY zeolite decreased upon heating because of its decomposition into NH₃ and H⁺. The amount of Au was closely related to that of NH₄⁺. In other words, before ion exchange, Au loading decreased to approximately zero on thermally treating the USY zeolite at 723 K, where NH₄⁺ was completely decomposed to yield the H-form zeolite. This indicated that NH₄⁺ played an important role in the deposition of Au on the zeolite. Formation of NH₄Cl (4 mol) was observed during the deposition of Au (1 mol) on the NH₄-USY zeolite, as verified by the IR, XRD, and temperature gravimetric analyses. This implied that the ion exchange of HAuCl₄ with NH₄⁺ took place simultaneously with the formation of NH₄Cl. The Au L₃-edge extended X-ray absorption fine-structure (EXAFS) analysis revealed the presence of Au₂O₃ on the as-prepared samples, which suggested that the hydrolysis of Au³⁺ proceeded to give Au₂O₃ on USY zeolites [19]. The maximum loading of Au on the NH₄⁺-USY zeolite was 5.5 wt%, which was obtained by employing an excessive amount of HAuCl₄. When 5.5 wt% Au was introduced into the NH₄⁺-USY zeolite, 1.1 mol kg⁻¹ of NH₄⁺ was eliminated to yield NH₄Cl, considering that the molar ratio of NH₄⁺ and Au was 4. The amount (1.1 mol kg⁻¹) corresponded to 70% of the total amount of NH₄⁺ (1.6 mol kg⁻¹) present in the NH₄⁺-USY zeolite. The efficient loading of Au on the Y-type zeolite may be due to the presence of a high concentration of NH₄⁺ (Brønsted acid sites) and a supercage having large pores that facilitated the reaction with HAuCl₄.

Figure 1. Dependence of NH₄⁺ concentration in ultrastable Y (USY) zeolites and Au loading on thermal treatment temperature of NH₄⁺-USY zeolites.

Figure 1. Dependence of NH₄⁺ concentration in ultrastable Y (USY) zeolites and Au loading on thermal treatment temperature of NH₄⁺-USY zeolites.

2.2. XRD Patterns Measured under Ex Situ and In Situ Conditions

Figure 2a shows the XRD patterns of a 3 wt%-Au/USY zeolite, which was first ion-exchanged with the NH₄⁺-USY zeolite and then treated at different temperatures in the atmospheres of 6% H₂ diluted with Ar. The 6%-H₂ treated samples were stored in the dark for over one month in air prior to collecting the XRD data. The peaks at 2θ = 38.2° and 44.4° were due to diffractions from the (111) and (200) planes of Au°, respectively. Weak peaks assignable to the diffractions of FAU zeolite appeared at 38.1° and 44.2°, which overlapped with the diffractions from Au°. The intensity of reflections from the Au°(111) and Au°(200) planes increased with an increase in the treatment temperature, i.e., from room temperature to 523 K. The appearance of steep reflections suggested the formation of aggregated crystals of Au metal. The intensity of these diffractions decreased and the peak became broader on further increasing the temperature, thereby suggesting a reduction in the particle size of Au. A marked reduction in peak intensity was observed at 573 K, indicating the formation of highly dispersed Au particles in the sample that was thermally treated at >573 K.

The data regarding the XRD patterns of the 3 wt%-Au/USY zeolites collected in an atmosphere of 3% H₂ under in situ conditions is depicted in Figure 2b. Broad diffractions due to the agglomerated Au particles were observed at ca. 38° and 44° in the XRD patterns of the Au/USY zeolites treated at 423–873 K. The change in the XRD patterns did not agree with those observed under ex situ conditions, i.e., the intensity of diffractions decreased at >573 K. In order to understand the contradiction between the data collected under in situ and ex situ conditions, the XRD patterns of the Au/USY zeolite stored in air were measured over a period of two months. Figure 3a shows the XRD patterns of the Au/USY zeolites treated at 473 K under a stream of 6% H₂ diluted with Ar, followed by storage in air at room temperature. The data was collected over a period of two months. Initially, the intensity of diffractions from Au°(111) and Au°(200) was small. The peaks gradually grew with time, and eventually, steep peaks were observed at 38.1° and 44.2° in the patterns of the Au/USY zeolites stored for 28–62 days. In contrast, no change was observed in the Au/USY zeolite treated at 773 K (Figure 3b). A significant difference between the samples treated at 473 and 773 K indicated that the temperature affected the change in the particle size of Au. Taking this into consideration, the appearance of steep peaks for the aggregated Au in Figure 2 (a, 423–523 K) was attributed to the growth of Au particles over a period of one month.

Figure 2. (a) XRD patterns of Au/USY zeolites treated at different temperatures under a 6% H₂ atmosphere diluted with Ar. Data was collected under ex situ conditions. Samples were stored in the dark for more than one month; (b) XRD patterns of Au/USY treated at different temperatures under a 3% H₂ atmosphere diluted with Ar. Data was collected under in situ conditions.

Figure 2. (a) XRD patterns of Au/USY zeolites treated at different temperatures under a 6% H₂ atmosphere diluted with Ar. Data was collected under ex situ conditions. Samples were stored in the dark for more than one month; (b) XRD patterns of Au/USY treated at different temperatures under a 3% H₂ atmosphere diluted with Ar. Data was collected under in situ conditions.

2.3. TEM Images of 3 wt%-Au/USY

TEM analysis was employed to observe the Au particles on the Au/USY zeolites. The TEM image of the as-prepared 3 wt%-Au/USY zeolite is shown in Figure 4a. Small particles were found to be distributed in high concentration over the zeolite support. Figure 4b shows that the size distribution of Au particles (presumably Au₂O₃ as found by XAFS [10]) was narrow, and the average diameter was 1.7 nm.

The TEM images and size distribution of the Au/USY zeolite treated at 473–1073 K under a 6% H₂ atmosphere are shown in Figure 5, Figure 6, respectively. The samples were stored in air for a month before conducting the TEM measurements. Large Au° particles of varying sizes were observed on the Au/USY treated at 473 K (Figure 5a). The shapes of the Au particles were not uniform. These observations agreed well with the XRD data in which steep reflections because of Au° appeared. In marked contrast, the Au° particles in the Au/USY treated at 773, 973 K, and 1073 K were much smaller and had homogeneous size distribution. It should be noted that the average size (1.8 nm) and distribution of the Au particles in the Au/USY treated at 773 K (Figure 6b) were close to those of the as-prepared Au/USY (Figure 4b). This suggested that the Au₂O₃ particles deposited on the NH₄⁺-USY zeolite were reduced by H₂ to yield Au° particles without significant migration or coalescence. Although the average size of the Au particles became slightly larger after the thermal treatment at 973 K (3.7 nm), their size distribution still remained narrow. The average size of the Au particles in the Au/USY treated at 1073 K was 6.3 nm, which was slightly larger than that of the particles treated at 773–973 K, and the distribution also became broader. Possibly, the pores of the USY zeolite partially collapsed owing to the formation of Au° because the size of the Au particles was larger than that of the particles present in the USY zeolites.

Figure 3. XRD patterns of Au/USY zeolites treated at (a) 473 K and (b) 773 K in a stream of 6% H₂, followed by storage under atmospheric conditions at room temperature.

Figure 3. XRD patterns of Au/USY zeolites treated at (a) 473 K and (b) 773 K in a stream of 6% H₂, followed by storage under atmospheric conditions at room temperature.

The mechanism of the coalescence of Au particles observed in the Au/USY zeolite treated at 473 K was not clear. One hypothesis was that the moisture present in air assisted the migration and growth of Au. In fact, the coalescence of Au nanoparticles was not observed in the 473 K-treated Au/USY zeolite when it was stored in dry air. Based on DFT calculations, it has been proposed that the coalescence of Au clusters could be prevented by anchoring them to the defective sites of SiO₂ [7]. Similarly, the OH groups generated on the defective sites in a moist atmosphere can possibly weaken the anchoring of Au on the USY support.

Figure 4. (a) TEM image and (b) particle size distribution of as-prepared 3 wt%-Au/USY zeolite.

Figure 4. (a) TEM image and (b) particle size distribution of as-prepared 3 wt%-Au/USY zeolite.

Figure 5. TEM images of Au/USY thermally treated in 6% H₂ diluted with Ar at (a) 473 K; (b) 773 K; (c) 973 K and (d) 1073 K. Samples were stored in the dark for more than one month.

Figure 5. TEM images of Au/USY thermally treated in 6% H₂ diluted with Ar at (a) 473 K; (b) 773 K; (c) 973 K and (d) 1073 K. Samples were stored in the dark for more than one month.

Figure 6. Distribution of Au particles in 3 wt% Au loaded on USY zeolites thermally treated in 6% H₂ diluted with Ar at (a) 473 K; (b) 773 K; (c) 973 K and (d) 1073 K.

Figure 6. Distribution of Au particles in 3 wt% Au loaded on USY zeolites thermally treated in 6% H₂ diluted with Ar at (a) 473 K; (b) 773 K; (c) 973 K and (d) 1073 K.

2.4. IR Study of Adsorbed CO

Low-temperature CO sorption experiments, monitored with the IR spectroscopy, were performed to determine the change in the dispersion of Au active sites by the thermal treatment temperature. It is well known that CO is a useful probe for Lewis acid sites and is characterized by the narrow and well-resolved bands in the region 2250–2000 cm⁻¹. Figure 7 shows FTIR spectra of USY (parent, spectrum a) and Au/USY treated at 773 K in 6% H₂ diluted with Ar (spectrum b) recorded after the exposure to CO. The interaction of CO with the USY zeolite resulted in the appearance of the 2178 cm⁻¹ band, assigned to CO when it is bonded to Brønsted acid sites (Si(OH)Al groups) [20]. The same band can be detected in the spectra of Au zeolites with different Au dispersions. Adsorption of CO to the Au species dispersed in the USY zeolite, which was hydrogen-reduced at 773 K, is manifested by the appearance of two bands at 2128 and 2092 cm⁻¹. The latter band (at 2092 cm⁻¹) can be attributed to metallic Au species, whereas the former to Au cationic species [21,22,23,24]. The origin of the 2128 cm⁻¹ band was confirmed in the CO sorption experiments, carried out on the samples exposed to O₂ at 773 K (spectrum c). It was evident that the contact of the reduced samples with oxygen transformed the metallic species (disappearance of the 2092 cm⁻¹ band) to cationic Au oxide clusters (the 2128 cm⁻¹ band grew). Additionally, the oxide form of the Au species in the oxygen-treated zeolite was confirmed by the presence of the 2340 cm⁻¹ band for CO₂ that resulted from the oxidation of CO by the Au oxide species.

Figure 8 shows the FTIR spectra of CO adsorbed on the USY (parent) (spectrum a) and the Au/USY zeolites treated at 473 K in 6% H₂ diluted with Ar (spectrum b). The sample was stored in the dark for more than one month before conducting the IR measurements. Both the metallic and oxide forms of the Au species were detected by the presence of CO. Nevertheless, the intensity of the CO band attributed to the metallic species (at 2097 cm⁻¹) was considerably lower (threefold) than that for the Au/USY treated at 773 K. The same effect was found for cationic Au oxide species, the presence of which was confirmed by the appearance of two bands at 2131 and 2123 cm⁻¹. These findings confirmed the distinctly lower dispersion of the Au species (metallic and oxide forms) in the Au/USY zeolite treated with H₂ at 473 K compared to that reduced at 773 K. Poorer activity of the Au species was confirmed by the relatively lower intensity of the CO₂ band at 2340 cm⁻¹: the oxidation of CO to CO₂ was better facilitated on the dispersed oxide species than on the aggregated ones.

Figure 7. FTIR of adsorbed CO on (a) parent USY, (b) Au/USY treated in 6% H₂ at 773 K and (c) Au/USY treated in 6% O₂ at 773 K.

Figure 7. FTIR of adsorbed CO on (a) parent USY, (b) Au/USY treated in 6% H₂ at 773 K and (c) Au/USY treated in 6% O₂ at 773 K.

Figure 8. FTIR of adsorbed CO on (a) parent USY, (b) Au/USY treated in 6% H₂ at 473 K, and (c) Au/USY treated in 6% O₂ at 473 K.

Figure 8. FTIR of adsorbed CO on (a) parent USY, (b) Au/USY treated in 6% H₂ at 473 K, and (c) Au/USY treated in 6% O₂ at 473 K.

2.5. Catalytic Performance in Aerobic Oxidation of Benzyl Alcohol

Partial oxidation of benzyl alcohol in air has been tested for Au/Y-type zeolites, because the catalytic activity of Au in the reaction was sensitive to the size of the Au particles [25]. Here, the main product was benzaldehyde while the by-product was a small amount of benzyl benzoate. Formation of benzaldehyde was negligible when the USY zeolites without Au loading were used. Figure 9a shows that the conversion of benzyl alcohol and the yield of benzaldehyde increased with an increase in the calcination temperature. The highest activity (conversion and yield) was obtained when the catalyst was treated at 673–973 K. Similarly, the dependence of the catalytic activity on the thermal treatment temperature was observed in the homocoupling of phenylboronic acid [19]. The change in the particle size of Au was correlated with the catalytic activity, which increased as the size of the Au particles decreased. Figure 10 shows the data of the repetitive reaction over the Au/USY zeolite. Only slight deactivation was observed even after repeating the reaction for more than 10 times. The data indicated that the Au/USY catalyst exhibited recyclability and could be reused at least 12 times without any loss of catalytic activity.

Figure 9b shows the conversion and yield of the Au/TiO₂ catalyst prepared by the deposition-precipitation method. The maximum activity was obtained after the catalyst was thermally treated in 6% H₂ diluted with Ar at 573 K. As a result, the conversion of benzyl alcohol and the yield of benzyladehyde were 71% and 54% respectively, which were lower than those obtained with the Au/USY zeolite treated at 673–973 K.

Figure 9. Dependence of conversion and yield of benzaldehyde on calcination temperature of (a) 3 wt%-Au/USY zeolite and (b) 3 wt%-Au/TiO₂ treated in a stream of 6% H₂ dilute with Ar. Closed symbol: benzyl alcohol conversion; Open symbol: benzaldehyde yield.

Figure 9. Dependence of conversion and yield of benzaldehyde on calcination temperature of (a) 3 wt%-Au/USY zeolite and (b) 3 wt%-Au/TiO₂ treated in a stream of 6% H₂ dilute with Ar. Closed symbol: benzyl alcohol conversion; Open symbol: benzaldehyde yield.

Figure 10. Catalytic performance of Au/USY zeolite treated at 773 K in 6% hydrogen in repetitive aerobic oxidation of benzyl alcohol.

Figure 10. Catalytic performance of Au/USY zeolite treated at 773 K in 6% hydrogen in repetitive aerobic oxidation of benzyl alcohol.

3. Experimental Section

3.1. Sample Preparation

Au was loaded on zeolites using HAuCl₄·4H₂O (Wako Pure Chemical Industries, Ltd., Tokyo, Japan) as a precursor. The NH₄-type USY zeolite (purchased from Tosoh, Tokyo, Japan), HSZ-341NHA, Si/Al₂ = 7.7, 1 g) was immersed in an aqueous solution of HAuCl₄ (250 mL, 6.1 × 10⁻⁴ mol L⁻¹), and the slurry was kept at 343 K for 1 h with constant stirring. The solution of HAuCl₄ was yellow but it decolorized during the treatment, while the zeolite became pale orange within 1 h of the solution reaching 343 K. The Au-loaded USY zeolite was filtered, washed with deionized water, and air dried in an oven at 323 K. The Au loading was measured by an ICP spectrometer after the digestion of Au/USY zeolites in aqua regia. The typical amount of Au loading was 3 wt%. The resulting Au/USY zeolite was subsequently treated with a stream of 6% H₂ diluted with Ar at a given temperature for 30 min. The temperature ramping rate was 5 K min⁻¹. Au loading on the CaNH₄⁺-type Y and NH₄⁺-type Y zeolites was carried out analogously to the preparation of Au/USY zeolites. The CaNH₄⁺-Y zeolite was prepared as reported elsewhere [26]. The concentration of Ca in the CaNH₄⁺-Y was 1.0 mol kg⁻¹. The ZSM-5 (JRC-Z5-90NA(1), Si/Al₂ = 90), mordenite (JRC-Z-HM15, Si/Al₂ = 15), and β-zeolites (JRC-Z-B25(1) Si/Al₂ = 25) were three times ion-exchanged with an aqueous solution of 0.5 M NH₄NO₃ at 343 K to yield NH₄⁺-form zeolites, prior to Au deposition. The deposition of Au was tried analogously on the NH₄⁺-mordenite, β-type, and ZSM-5 zeolites for loading Au on Y-type zeolites. The precipitation-deposition method was employed for loading 3 wt% Au on TiO₂ (JRC-TIO-11, supplied by the Catalysis Society of Japan). First, 0.5 g of zeolite was added to a 100 mL solution of HAuCl₄ (Wako Pure Chemical Industries, Ltd., Tokyo, Japan) heated in a water bath at 343 K. The pH of the solution was adjusted to 6 by adding an ammonia solution (2.8%) with vigorous stirring. After adjusting the pH, the slurry was stirred for an additional 1 h. After filtration, the Au-loaded sample was thoroughly washed with deionized water until no Cl⁻ was detected with AgNO₃ solution. Finally, the sample was dried in air at 323 K for 3 h.

3.2. TEM and XRD Data Collection and Analysis

The TEM images were obtained using a HITACHI H-9000UHR microscope with an acceleration voltage of 300 kV. The crystalline structure of the Au zeolite was analyzed by XRD under ambient conditions using a Rigaku Ultima IV X-ray diffractometer with Cu Kα radiation. In situ XRD data was obtained using RIGAKU SmartLab equipped with a Johannson-type primary monochromator in the atmosphere of 3% H₂ diluted with Ar. The measurement was carried out in Bragg-Brentano geometry. Temperature ramping rate and scan speed were 5 K min⁻¹ and 4° min⁻¹, respectively.

3.3. IR Measurements and Analysis

Prior to IR studies, the 3 wt%-Au/USY zeolites were pressed into thin discs (5 mg/cm²) and thermally treated in situ in an IR cell under vacuum at 770 K for 1 h. Sorption of CO (Linde Gas Krakow, Poland 99.95%) was carried out to detect all types of reduced Au species. After collecting data for the reduced samples, the Au species were oxidized by bringing zeolites in contact with oxygen (obtained in situ by the thermal decomposition of KMnO₄) at either 773 K or 473 K (for respective samples) for 1 h. At this stage, the sorption of CO doses allowed us to discriminate between the oxidized and reduced species. In each experiment, CO was sorbed at 173 K until all the Au species were saturated (maximum intensities of the bands were observed at ca. 2130 and 2090 cm⁻¹). Additionally, the appearance of the 2178 cm⁻¹ band corresponding to the Si(OH)Al groups pointed to the complete saturation of all Au species. The IR spectra were recorded with a Bruker Tensor 27 spectrometer (equipped with an MCT detector) with a spectral resolution of 2 cm⁻¹.

3.4.Catalytic Reactions

Partial oxidation of benzyl alcohol was carried out over Au/USY zeolites in air. The samples were pretreated in 6% H₂ at a given temperature before being used in the reaction. The treated samples were stored in the dark. Benzyl alcohol (1.0 mmol; Wako Pure Chemical Industries, Tokyo, Japan), K₂CO₃ (1.0 mmol; Wako Pure Chemical Industries, Tokyo, Japan), toluene (5 mL; Wako Pure Chemical Industries, Tokyo, Japan), and Au-loaded catalysts (10 mg) were used. The molar amount of Au was 1 mol% with respect to benzyl alcohol. A sample bottle (20 mL) was placed on a magnetic stirrer and subjected to vigorous stirring. The reaction was performed at 323 K under atmospheric conditions for 8 h. The reaction was quenched by adding acetone to the solution, and the solution was analyzed by Shimadzu 2010 Gas Chromatograph equipped with an InertCap Pure-WAX (30 m) capillary column. Tridecane was used as the internal standard for the quantitative analysis. For recycling tests, a fresh substrate was added to the catalyst after its precipitation at the bottom of the reactor and removal of the supernatant. The reaction was repeated under the aforementioned conditions.

4. Conclusions

Au was successfully loaded on the NH₄-form Y-type zeolite to yield 1.7-nm-sized Au₂O₃ along with the formation of NH₄Cl. The particle size of Au° generated through the thermal treatment of the Au₂O₃/USY zeolite in 6% H₂ diluted with Ar was highly dependent on the treatment temperature. The TEM analysis ascertained that the Au/USY zeolite treated at 473 K with subsequent preservation in air for one month resulted in the formation of 25-nm-sized Au particles. In contrast, formation of 1.8–3.7 nm Au particles was observed after thermal treatment at 773–973 K. This behavior agreed with the results of the XRD and IR analyses of the adsorbed CO. The intensity of the CO band, attributed to the metallic species in the Au/USY-773 K zeolite, was considerably higher (threefold) than that observed in the Au/USY-473 K zeolite. Moreover, the catalytic activity of the Au/USY zeolites in the partial oxidation of benzyl alcohol showed good correlation with the Au dispersion. The highest activity was attained with the Au/USY treated at 773–973 K. The simplicity of the technique used for Au loading is worth emphasizing: stirring a solution of HAuCl₄ and NH₄-type USY zeolites at 343 K afforded Au₂O₃. This study showed the possibility of regulating the size of Au particles by employing USY zeolites.