Size dependence of silver nanoparticles in carboxylic acid functionalized mesoporous silica SBA-15 for catalytic reduction of 4-nitrophenol

Abstract

In this study, the formation of silver nanoparticles (Ag NPs) with a particle size of about 3 nm was successfully achieved by using the mesoporous channels of carboxylic acid (–COOH) functionalized SBA-15 as the support. When pure silica SBA-15 (without –COOH groups) was employed, the Ag NPs were formed outside of the mesopore and their particle size was significantly larger, up to about 20 nm. The –COO⁻ groups under basic conditions can effectively interact with the Ag⁺ ions, and thus allowed facile fabrication of Ag NPs. The catalytic activity of these Ag NPs based SBA-15 materials were tested for a model reaction, namely the reduction of 4-nitrophenol to 4-aminophenol. The results showed that the particle size of the Ag NPs play a key role in determining their catalytic activity. The apparent kinetic constant for the Ag NPs with a particle size of about 3 nm was 1.1 × 10⁻² s⁻¹, corresponding to the activity parameters of 12.2 and 4630 s⁻¹ g⁻¹ by considering the total mass of the catalyst used and the Ag NPs alone, respectively, which were remarkably high as compared to other matrices bearing Ag. Moreover, the Ag NP based materials exhibited good recyclability up to 5 cycles.

---

Introduction

Considerable research effort has been devoted on the fabrication of metal nanoparticles (NPs) because of their physicochemical properties, which are unique and different from their bulk forms. Metal nanoparticles have gained significant interest due to their promising applications in electronics, chemical, biological and catalysis.^1–4 In particular, the advantageous properties of a large surface area to volume ratio of NPs makes them highly catalytically active in a variety of chemical reactions. As a result, much effort has been dedicated to develop different synthesis methodologies for fabrication of NPs with various shapes. Roucoux et al. recently reported about the different methods of metal nanoparticles synthesis and their catalytic applications.⁵ However, the main drawback of metal nanoparticles as catalysts is their tendency to aggregation, which often results in uncontrollable and disadvantageous effects on their activity and selectivity. In order to achieve metal nanoparticles with controllable particle size, shape and distribution, an appropriate support to stabilize the metal nanoparticles is often needed for their practical applications. In the past few years, a variety of protective systems, for example, ionic compounds,⁶ ionic surfactant,⁷ ligands,^8,9 polymer based materials,^10,11 mesopolymer FDU-15,¹² ordered mesoporous silicas,^13,14 polymer micelles,¹⁵ carboxylate modified polyvinylpyrrolidone (PVP),^16–18 hydrogels,¹⁹ mesoporous carbons,^20,21 graphene oxide,^22,23 mesoporous cellular foams,²⁴ and fibrous nano-silica²⁵ has been tested to overcome the aggregation problem associated with the fabrication of metal nanoparticles. Among them, mesoporous silica materials turn out to be suitable matrices for fabrication of metal nanoparticles because their unique structural characteristics, for example, high surface area, large pore volume, and tunable pore size, are advantageous to the nanoparticles dispersion, and thus can avoid the aggregation problem of NPs. For example, metal nanoparticles and nanowires have been successfully synthesized within the channels of mesoporous silicas MCM-41 and SBA-15.^26–29 However, it is not an easy job to control the growth of nanoparticles within the mesopores since the interactions between the surface silanol groups and the metal precursors are relatively weak. In this situation, additional reducing agent or stabilizer is needed for fabrication of metal nanostructures. For example, when Ag⁺ ions interact with negatively charged SiO₂ surface and then followed by their reduction with trisodiumcitrate, different Ag nanoparticles and nanorods have been successfully formed inside the SBA-15 mesoporous channels.³⁰ Thus, additional surface modifications of mesoporous silicas are desirable to endow appropriate interactions between the support surfaces and the metal precursors. In fact, the mesopore surfaces of mesoporous silicas have been modified by a variety of organic functionalities, such as amine,^31,32 thiol,³³ and carboxylic acid,^34,35 in order to synthesize metal nanoparticles. However, it was found that the distribution of the organic functional groups, either on the pore surface or inside the mesopore, played a key factor in determining the size control of metal nanoparticles.

Silver nanoparticles (Ag NPs) have emerged as promising candidates for applications in some active research fields such as biomedicines, catalysis, energy conversion, and photoelectronic devices.^36–40 In addition, silver (Ag) is relatively cheap and environmentally friendly in comparison to other noble metals such as Au, Pt, and Pd. As to the applications in catalysis, Ag NPs have been demonstrated remarkable performance in several catalytic reactions, such as methanol and ethanol oxidation conversions, butadiene epoxidation, oxidation of CO and selective NO_x reduction.^41–45 Moreover, a model reaction, that is, reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP) has been frequently tested for the catalytic activity of Ag NPs within the different forms of matrices.^12,25,30 4-NP is highly soluble and stable in water, and thus it is difficult to decompose. Consequently, it becomes a well-known industrial pollutant and source of great environmental concern.^46,47 Therefore, an effective methodology for its removal is highly desirable. Although a variety of processes has been developed for its disposal, the most economical and efficient approach is to reduce the nitro group to amino group.^48–51 Besides, the reduction product 4-AP can be reused since it is a key intermediate when drugs, dyes, corrosion inhibitors, and photographic developers are processed.^52,53 Therefore, it is of strong need to develop an efficient catalyst for the reduction of 4-NP.

In the present work, mesoporous silicas SBA-15 functionalized with and without –COOH functional groups were used as the supports to fabricate Ag NPs embedded SBA-15 based nanocatalysts. The presence of the –COOH moiety in the mesopores of SBA-15 not only facilitates more Ag⁺ ions that can enter the mesopores through the electrostatic interaction between –COO⁻ groups and Ag⁺ ions under basic conditions, but also is favorable for the formation of Ag NPs with smaller sizes (down to around 3 nm). These nanocatalysts were characterized and demonstrated to have excellent catalytic activity in the catalytic reaction of 4-NP to 4-AP.

Experimental section

Materials

Tetraethyl orthosilicate (TEOS), triblock copolymer Pluronic P123 (EO₂₀PO₇₀EO₂₀), and silver nitrate (AgNO₃) were received from Sigma-Aldrich. Carboxyethylsilanetriol sodium salt (abbreviated as CES, 25 wt% in water) was purchased from Gelest. 4-NP was purchased from Sigma-Aldrich. Sodium borohydride (NaBH₄) was purchased from Tianjing Guangfu Chemical Company. All chemicals were used as received without additional purification procedures.

Preparation of Ag NPs in SBA-15

The synthesis of mesoporous silicas SBA-15 with and without –COOH functional groups was followed the procedures that we reported previously.⁵⁴ Briefly, 1 g of Pluronic P123 was dissolved in 31.25 g of 1.9 M HCl solution and continuously stirred at 40 °C for 4 h. Then, the premixed TEOS and CES were added dropwise to the solution and stirred vigorously at 40 °C for 20 h. The reaction mixture was hydrothermally treated at 100 °C for 24 h. The composition of the reaction mixture was varied in the range of x/100CES : (100 − x)/100TEOS : 0.0168P123 : 5.85HCl : 162.68H₂O. The amount of CES used in this study was x = 10. The final precipitate was filtered off, washed with water and air-dried at room temperature. To remove the polymer template, 0.5 g of the as-synthesized sample was dispersed in 60 mL of 48 wt% H₂SO₄ solution, and the mixture was heated at 95 °C for 24 h. Finally, the sample was recovered by washing with water and dried at 90 °C. The resultant SBA-15 samples were denoted as SBA-15-(x), where x is either 0, representing pure silica SBA-15, or 10, which is the molar ratio of CES/(TEOS + CES) of 10%. For the synthesis of Ag nanoparticles by using SBA-15-(x) as a support, 0.03 M AgNO₃ with different volumes (10, 20, 30, 50 mL) was wet impregnated in 0.1 g of SBA-15-(x) at room temperature with vigorous stirring for 3 h. A certain amount of 0.1 M NaOH_(aq) was added to control the pH of the solution to be 9. Then the sample was filtrated, washed with deionized water, and then dried at room temperature. Afterward, the sample was thermally treated at 300 °C, with a heating rate of 1 °C min⁻¹, for 2 h under a flow of Ar/H₂ (95%/5%). This process helps to retain some of the –COOH groups that are pendant on the mesopore surface. Meanwhile, the presence of –COOH groups stabilizes the Ag⁺ ions and thus plays a key role in the in situ formation of Ag nanoparticles with a controllable size. The samples were denoted as Ag(y)@SBA-15-(x), where y is the amount (mmol) of AgNO₃ used in the synthesis and x denotes the molar percentage ratio of CES/(TEOS + CES). The synthesis procedures are outlined in Scheme 1.

Scheme 1 Schematic illustration of the synthesis of Ag(y)@SBA-15-(x) nanocatalysts.

Characterization methods

Powder X-ray diffraction (XRD) measurements were performed on Wiggler-A beamline (λ = 0.1321712 nm) provided by NSRRC (National Synchrotron Radiation Research Center) in Taiwan. The textural properties of the samples investigated in this study were obtained by the analysis of the N₂ adsorption–desorption isotherms, which were measured on a Quantachrome Autosorb iQ-2 analyzer. The relative pressure range of P/P₀ = 0.05–0.3 was selected to obtain specific surface areas by using the Brunauer–Emmett–Teller (BET) method. On the other hand, the volumes of N₂ adsorbed at P/P₀ = 0.95 or in the vicinity were used to determine pore volumes. The primary mesopore volume, micropore volume and external surface area were calculated using the α_s plot method from the data in the range from 1.4 to 2. The pore size was calculated using an equation derived on the basis of geometrical considerations of the structure of the material. The pore diameter, W_d, was measured on the basis of the XRD (100) interplanar spacing, d₁₀₀, the volume of primary mesopores, V_p, and the volume of micropores, V_mi, by following the relation⁵⁵where c is a constant and its value is dependent on pore geometry and is equal to 1.213 for cylindrical pores. ρ is the density of pore walls and assumed to be 2.2 g cm⁻³.

Transmission electron microscopic (TEM) images were taken on a JEOL JEM2010 microscope. The Ag contents of the samples were determined by Inductively-Coupled Plasma Atomic Emission Spectroscopy (ICP-AES, Jarrell-Ash, ICAP 9000). Fourier transform infrared (FTIR) measurements were conducted on a JASCO 4200 spectrometer. Ultraviolet-visible (UV-Vis) absorption spectra were recorded by a PG Instrument T90+ UV-Vis double beam spectrophotometer. Thermogravimetric analysis (TGA) was performed using a Perkin-Elmer TGA 7 instrument at a heating rate of 10 °C min⁻¹ under a nitrogen flow. X-ray photoelectron spectroscopy (XPS) was employed to determine the state of Ag NPs by using a Thermo VG Scientific Sigma Probe spectrometer, which has a monochromated Al K_α radiation with a spot size of 400 μm.

Catalytic studies

First, a 0.09 mM aqueous solution containing 4-nitrophenol (22.5 mL) was added to 0.0306 M freshly prepared NaBH₄ solution (5 mL). A deep yellow solution was formed after addition. Subsequently, an aqueous mixture consisting of 0.9 mg Ag(y)@SBA-15-(x) catalyst and 12 mL H₂O was added to the above yellow solution. A small portion of the solution was immediately put into a quartz cuvette for UV-Vis absorption measurements after the solutions were mixed thoroughly. When the reaction was complete, a colorless solution was obtained. The intensity change of the absorption peak of the reaction mixture as a function of reaction time was indication of the progress of the reaction. The spectral range measured in a UV-Vis spectrophotometer was in the range of 250–500 nm.

Reusability of the catalyst

The reuse property of the catalyst was analyzed by a series of recycle catalysis experiments. After the first cycle of reaction, the Ag-NPs embedded SBA-15-(x) was separated by centrifugation, washed with 50 mL distilled water at least 2–3 times, and finally dried under vacuum at 40 °C. In the further cycles of reaction, the dried catalyst was repeatedly used according to the above mentioned procedures.

Results and discussion

Characterization of catalysts

The small angle XRD patterns (Fig. 1A) of Ag(y)@SBA-15-(x) samples show three diffraction peaks in the region of 2θ = 0.8–2.0°, which corresponded to the (100), (110) and (220) planes that are characteristics of mesoporous materials with the 2D hexagonal pore structure.⁵⁶ As shown in Fig. 1A, the intensity of these diffraction peaks decreases with the increase in the volume of the AgNO₃ solution. It suggests that the increased amount of Ag nanoparticles formed can either fill or block the pores, which reduce the electronic contrast between the silica walls and the pores and thus decrease the intensity of diffraction peaks.

Fig. 1 (A) Small angle and (B) wide angle XRD patterns of (a) Ag(0.3)@SBA-15-(0) and Ag(y)@SBA-15-(10), where y = (b) 0.3, (c) 0.6, (d) 0.9, and (e) 1.5.

The wide angle diffraction patterns of the samples can further support the presence of Ag nanoparticles. As shown in Fig. 1B, the presence of several diffraction peaks at 2θ values of about 38.3°, 44.5°, and 64.6° are clearly observable, which correspond to (111), (200), and (220) planes, respectively, of face-centered cubic (fcc) structure of Ag nanoparticles with lattice constant of a = 4.086 Å (Fig. 1B) [JCPDS no. 04-0783].^25,57 These indexed peaks clearly demonstrate the crystalline nature of Ag NPs within the mesoporous silica matrix. The average crystallite size D (nm) of Ag NPs can be obtained according to the Scherrer equation,⁵⁸

where K_s is a dimensionless shape factor with a value close to unity, λ is the X-ray wavelength, which is 0.15405 for Cu K_α, B is the full width at the half-maximum (FWHM) of the peak in radians and θ is the diffraction angle. According to the analysis results, the minimum average particle size of Ag NPs within Ag(0.3)@SBA-15-(10) was found to be 2.4 nm, and the size was increased to 16–22 nm when more Ag⁺ ions were present in the initial mixture. Without the presence of the –COOH moiety on the mesopore surface, the particle size of Ag NPs within pure silica SBA-15 (i.e., SBA-15-(0)) was significantly larger (18.3 nm) under the conditions of the same concentration of Ag⁺ ions (e.g., 0.3 mmol). Based on the fact that the Ag(0.3)@SBA-15-(10) sample possesses smaller sizes of Ag NPs, it implies that the –COOH functionalization of SBA-15 plays a key factor to control the size of the Ag NPs during the formation process.

Fig. 2 displays the nitrogen adsorption–desorption isotherms of the Ag(y)@SBA-15-(x) samples. The isotherms of the silica supports SBA-15-(0) and SBA-15-(10) (i.e., parts (a) and (b) in Fig. 2) are of type IV with H1 hysteresis loops. When Ag⁺ ions were loaded on SBA-15-(0), the isotherm (part (c) in Fig. 2) almost retained the same characteristics for Ag(0.3)@SBA-15-(0). This observation suggested that the formed Ag NPs were mostly located on the external surface of SBA-15-(0). As shown in parts (d) and (e) in Fig. 2, when SBA-15-(10) was used as the support for formation of Ag NPs, on the other hand, the hysteresis loops shifted to lower relative pressures (P/P₀), and were not as steep as that of SBA-15-(10), indicating smaller mesopore widths along with broader pore size distributions due to pore constriction by the occupancy of Ag NPs in the mesopores. Table 1 summaries the textural properties analyzed from the isotherms. As seen in Table 1, a significant decrease in the surface area and pore volume of Ag(y)@SBA-15-(x) was observed when the initial amount of Ag⁺ ions was increased, which could be due to the blockage of some pores by the Ag NPs. In addition, the surface areas of Ag(0.3)@SBA-15-(x) (x = 0 and 10) nanocatalysts are 425 and 496 m² g⁻¹ (Table 1), respectively, indicating that both nanocatalysts exhibit relatively high surface areas for their catalytic activity.

Fig. 2 N₂ adsorption–desorption isotherms of Ag(y)@SBA-15-(x), where (y, x) = (a) (0, 0), (b) (0, 10), (c) (0.3, 0), (d) (0.3, 10), (e) (0.6, 10), and (f) (1.5, 10).

Table 1 Structural and textural properties of the samples studied

Sample	d100 (nm)	a0^a (nm)	ABET^b (m^2 g^−1)	Vt^c (cm^3 g^−1)	Wd^d (nm)	Average Ag NPs size^e (nm)	Ag^f (wt%)
a a0 is the cell parameter, .b ABET is the surface area.c Vt is the total pore volume.d Pore diameters (Wd) were calculated from the geometrical relation given in eqn (1).e Average Ag NPs particle size estimated from the XRD patterns using Scherrer formula.f Amount of Ag determined by ICP-AES.
SBA-15-(0)	10.7	12.4	787	1.05	10.0	—	—
Ag(0.3)@SBA-15-(0)	9.5	11.0	425	0.79	6.8	18.3	0.205
SBA-15-(10)	10.5	12.1	691	1.05	9.4	—	—
Ag(0.3)@SBA-15-(10)	9.9	11.4	496	0.71	8.4	2.4	0.264
Ag(0.6)@SBA-15-(10)	9.3	10.7	392	0.55	7.5	20.4	0.272
Ag(0.9)@SBA-15-(10)	9.5	11.0	255	0.36	5.9	16.6	0.305
Ag(1.5)@SBA-15-(10)	9.7	11.2	184	0.25	4.5	22.0	0.316

---

The ICP-AES analysis (Table 1) indicated that the real content of Ag NPs loaded on SBA-15-(10) can be increased by gradually increasing the Ag⁺ concentration in the initial reaction mixture. It should be noted that the carboxylic acid groups can have electrostatic interaction with the metal precursor, which not only governs the size and dispersion of the nanoparticles, but also facilitate the Ag⁺ reduction within the channels of SBA-15.

TEM was employed to further confirm the formation of Ag NPs and their particle size distribution within the mesopores of SBA-15. As seen in Fig. 3, the Ag NPs are uniformly dispersed in the SBA-15-(10) with a size of around 3 nm. These Ag NPs are found to be spherical in shape. It should be noted that neither additional reducing agent such as citrate nor stabilizer was needed to prepare such a small sized Ag NP in the present study. The energy dispersive X-ray spectrum (EDS) from the corresponding TEM image of the Ag(0.3)@SBA-15-(10) sample is shown in Fig. S1 (ESI†). It is clear from the spectrum that the Ag(0.3)@SBA-15-(10) nanocatalyst consists of C, O, Si and Ag elements. The Cu peak comes from the copper grid. When higher Ag⁺ concentrations were used, the small size Ag NPs cannot formed inside the SBA-15-(10). Instead, the Ag NPs with the particle size around 20 nm are formed. Without the stabilization of –COOH groups, on the other hand, the particle size of Ag NPs on SBA-15-(0) is significantly larger, around 22 to 35 nm, which is much larger than the pore size of SBA-15. Therefore, these Ag NPs are located outside the mesopores.

Fig. 3 TEM images of Ag(y)@SBA-15-(x), where (y, x) = (a) (0.3, 0), (b) (0.3, 10), and (c) (1.5, 10).

In order to investigate the role of carboxylic acid groups, FTIR was employed to reveal the interactions between the silver ions and the –COOH groups. The FTIR spectra of Ag(y)@SBA-15-(x) nanocatalysts are displayed in Fig. S2 (ESI).† Without introducing the Ag⁺, a typical band at 1722 cm⁻¹ was clearly visible for the support SBA-15-(10), which indicated the successful incorporation of carboxylic acid groups in SBA-15. After the fabrication of Ag NPs, the intensity of the band at 1722 cm⁻¹ decreased significantly, which might be due to the calcinations at high temperatures. Still, weak band at higher Ag⁺ concentrations confirmed the presence of some –COOH groups. The shifting of the band to 1731 cm⁻¹ suggests some interactions between the –COOH groups and the Ag NPs. The absorption band observed for SBA-15-(10) sample at ca. 958 cm⁻¹ was attributed to stretching vibration of Si–OH groups. The intensity of the band gradually decreased for the Ag(y)@SBA-15-(10) nanocatalysts with increased Ag⁺ concentration, suggesting some changes in the interaction with Si–OH groups. As reported previously, this interaction could be assigned to stretching vibration of the Si–O–Me linkage (where Me = Ag in this case).^59,60 This finding also suggested the Ag NPs were formed in the mesoporous silica framework of SBA-15.

TGA was carried out to investigate the thermal stability of catalysts. The TGA curves for the template extracted SBA-15-(10) and Ag(0.3)@SBA-15-(10) samples are shown in Fig. S3 (ESI).† The preliminary weight loss of around 1–2 wt% below 200 °C was due to the physisorbed water. The weight loss started at 240 °C, which was attributed to the decomposition of some –COOH groups. Further weight losses of around 6 and 10 wt% were observed in the temperature range of 400–570 °C for SBA-15-(10) and Ag(0.3)@SBA-15-(10), respectively, due to the decomposition of the –COOH groups and condensation of silica walls. The result suggests that a certain amount of –COOH groups are present for in situ formation of Ag NPs, which is also observed in the FTIR analysis.

XPS measurements were conducted in order to explore the chemical states associated with the Ag NPs embedded in the mesoporous matrix. Fig. 4 shows the XPS data of Ag(0.3)@SBA-15-(0) and Ag(0.3)@SBA-15-(10) nanocatalysts in the binding energy regions of 364 to 380 eV, which are related to silver species. As shown in Fig. 4, the Ag 3d_5/2 and 3d_3/2 spin–orbit coupled core levels were observed at 368.8 eV and 374.9 eV, respectively, for Ag(0.3)@SBA-15-(10) sample. For the Ag(0.3)@SBA-15-(0) sample, the Ag 3d_5/2 and 3d_3/2 peaks were observed at 368.3 eV and 374.2 eV, respectively. The difference in spin energy levels of 6.1 eV and 5.9 eV for Ag(0.3)@SBA-15-(10) and Ag(0.3)@SBA-15-(0), respectively, implied the successful formation of the metallic Ag NPs (Ag⁰) within the channels of SBA-15.⁶¹ Normally, the Ag 3d_5/2 peak is at 368 eV for bulk and Ag 3d_3/2 6 eV higher. In comparison to bulk silver, the shifting of the peaks by 0.3 eV to higher binding energies in Ag(0.3)@SBA-15-(0) could be due to the interaction between the Si–OH groups of SiO₂ support and the Ag NPs. In the case of Ag(0.3)@SBA-15-(10), the peaks shifted by 0.9 eV to higher binding energies. The higher value of peak shifting observed in Ag(0.3)@SBA-15-(10) suggests that there may be additional stronger interactions between the –COOH groups of functionalized SBA-15 and the Ag NPs in comparison with that of pure silica SBA-15. Therefore, the residual –COOH groups play a definitive role in the formation of Ag NPs for the Ag(0.3)@SBA-15-(10) nanocatalyst.

Fig. 4 Ag 3d XPS data of (a) Ag(0.3)@SBA-15-(0) and (b) Ag(0.3)@SBA-15-(10) nanocatalysts.

Catalytic activity

The reduction conversion of 4-NP to 4-AP in the presence of NaBH₄ was employed as a model reaction to evaluate the catalytic efficiency of the Ag NPs in SBA-15. It has been demonstrated that the reaction was a thermodynamically favorable process but proceeded very slowly in the absence of a catalyst. On the other hand, the reaction time can be significantly shortened when the Ag NPs were served as the catalysts. Normally, an aqueous solution of 4-NP gives a maximum absorption band in the UV-Vis spectral region at 317 nm, which has a red shift to 400 nm after immediate addition of NaBH₄ owing to the formation of 4-nitrophenolate ion (part (a) and (b) in Fig. S4, ESI†).⁶² Simultaneously, the color of the solution changes from light-yellow to yellow-green. The position of the maximum absorbance at 400 nm did not shift over time even after the addition of the NaBH₄ solution, indicating that the presence of NaBH₄ did not give rise to the reduction reaction. Moreover, no reaction occurred when the support SBA-15-(x), (x = 0 or 10) was added to the mixed solution of 4-NP and NaBH₄, since the intensity of UV absorbance did not change (part (c) in Fig. S4, ESI† and part (a) in Fig. 5). It revealed that both the SBA-15-(x) (x = 0 and 10) supports were inactive with respect to the reduction reaction of 4-NP.

Fig. 5 UV-Vis spectra of the successive reduction of 4-NP over (a) Ag(0)@SBA-15-(10), (b) Ag(0.3)@SBA-15-(0), (c) Ag(0.3)@SBA-15-(10), and (d) Ag(1.5)@SBA-15-(10).

Upon addition of a small amount of Ag NPs catalysts (0.9 mg) to the reaction medium, there is a gradual decrease in the intensity of absorption peak at 400 nm as a function of time, indicative of the reduction of 4-NP. Meanwhile, the amount of 4-AP increased with increasing reaction times gave rise to a concomitant increase of a new band at 298 nm, as shown in Fig. 5. The degree of the reaction can be monitored and followed by measuring the intensity changes in the corresponding UV-Vis absorption bands.

The completion of the reaction of 4-NP to 4-AP was indicated by the appearance of a colorless solution, which was within 10 to 15 min in the present case. It was proposed that both 4-NP and BH₄⁻ ions were adsorbed on the surface of Ag(y)@SBA-15-(x) in the first stage.⁶³ Then, the BH₄⁻ ion donated electrons to 4-NP as the acceptor via Ag NPs. After the electrons were transferred to the Ag NPs, the hydrogen atoms formed from NaBH₄ reduced 4-NP molecules, which can be confirmed visually by watching the color change in solution from yellow to colorless. Since the NaBH₄ concentration is very high as compared to 4-NP, it can be considered as a constant during the course of the reaction. Under such conditions, a pseudo first-order, with respect to 4-NP concentration, for the reduction reaction can be assumed. On the basis of these assumptions, the reaction conversion can be directly calculated from C_t/C₀, where C₀ and C_t are the concentrations of 4-NP at the beginning and at the reaction time t. The C_t/C₀ ratio was in turn determined by measuring the relative intensity of UV-Vis absorbance (A_t/A₀) at 400 nm. As shown in Fig. 6, the linear dependence of ln(C_t/C₀) versus the reaction time clearly indicated that the reaction followed the pseudo first-order behavior. Thus, with ln(C_t/C₀) = −kt, where k is the apparent first-order rate constant (s⁻¹), and t is the reaction time, the reaction rate constant k was calculated from the slopes of the plots and found to be 1.1 × 10⁻² s⁻¹ for Ag(0.3)@SBA-15-(10), 2.3 × 10⁻³ s⁻¹ for Ag(1.5)@SBA-15-(10), but only 3.4 × 10⁻³ s⁻¹ for Ag(0.3)@SBA-15-(0). The correlation coefficient of the linear plot was as high as 0.997. It was observed that Ag(0.3)@SBA-15-(10) exhibited the highest catalytic activity as compared to other samples used in this study. This observation could be directly related to the particle size parameters of Ag NPs. As reported in the literature, smaller nanoparticles have tendency to exhibit higher catalytic activity since they have the property of greater surface-to-volume ratio.⁶⁴

Fig. 6 Plots of ln(C_t/C₀) versus reaction time for the reduction of 4-NP over (a) Ag(0)@SBA-15-(10), (b) Ag(0.3)@SBA-15-(0), (c) Ag(0.3)@SBA-15-(10), and (d) Ag(1.5)@SBA-15-(10).

The activity parameter κ (= k/M, where M is the total mass of the catalyst used) was calculated for the purpose of quantitative comparison.^65,66 Thus, the activity parameters of 12.2, 2.56, and 3.78 s⁻¹ g⁻¹ were obtained for the reduction of 4-NP by using Ag(0.3)@SBA-15-(10), Ag(1.5)@SBA-15-(10), and Ag(0.3)@SBA-15-(0), respectively. It is worth noting that the actual catalyst is the Ag NPs alone since there is no activity for the support SBA-15-(10), as shown in part (a) of Fig. 5. Based on the weight percentage (0.264 wt%, Table 2) of Ag NPs in Ag(0.3)@SBA-15-(10), the true activity parameter turned out to be 4630 s⁻¹ g⁻¹ if only the amount of Ag NPs was considered. This value is remarkably high as compared to the values for the catalysts reported in the literature on the same basis of the content of NPs only, instead of the whole catalyst.^24,67–69 For example, some Ag NPs based catalysts such as AgNPs/silica nanotube and MCF-100-Ag-0.01 only exhibited the activity parameter of 142.26 s⁻¹ g⁻¹ and 204.62 s⁻¹ g⁻¹, respectively.^24,67 For Pt–Au alloy based RGO catalysts, the activity parameter of 926 s⁻¹ g⁻¹ and 1700 s⁻¹ g⁻¹ has been reported for Pt–Au pNDs/RGOs and Pt3Au1-PDA/RGO, respectively.^68,69 All these catalysts showed lower activity parameters than that of Ag(0.3)@SBA-15-(10). Similarly, the activity parameters for Ag(1.5)@SBA-15-(10), and Ag(0.3)@SBA-15-(0) were found to be 809 and 1843 s⁻¹ g⁻¹, respectively. Table 2 shows the detailed results of activity parameters with or without consideration of the catalyst support SBA-15.

Table 2 Catalytic activities of Ag(y)@SBA-15-(x) nanocatalysts for the reduction of 4-NP

Catalyst	Amount used for catalysis (mg)	AgNPs loading^a (wt%)	k (s^−1)	k/M^b (s^−1 g^−1)	k/M^c (s^−1 g^−1)
a The value was determined by ICP-AES.b Considering the total mass of the catalyst.c Considering only the mass of the Ag amount in the catalyst.
Ag(0.3)@SBA-15-(0)	0.9	0.205	3.4 × 10^−3	3.78	1843
Ag(0.3)@SBA-15-(10)	0.9	0.264	1.1 × 10^−2	12.2	4630
Ag(1.5)@SBA-15-(10)	0.9	0.316	2.3 × 10^−3	2.56	809

---

A comprehensive comparison of the activity parameters with the consideration of total mass of the catalyst in the reduction of 4-NP for Ag NPs based catalysts is also given in Table 3. According to the literature, the ordered mesoporous silver nanoparticle/carbon composite (Ag/C-0.05) exhibited the activity parameter of 2.66 s⁻¹ g⁻¹.⁵⁷ Tang et al. fabricated silver nanoparticle doped submicron carbon spheres, i.e., Ag-NP/C composite, and found the activity parameter value to be 1.69 s⁻¹ g⁻¹.⁶¹ The TAC-Ag-1 catalyst⁷⁰ and core–shell structured Fe₃O₄@SiO₂–Ag magnetic nanocomposite⁷¹ have the activity parameters of 1.30 and 7.67 s⁻¹ g⁻¹, respectively. A maximum activity parameter of 4.03 s⁻¹ g⁻¹ was given by Ji et al. for their mono-dispersed Ag nanoparticles decorated macrotube/mesopore carbon catalysts.²¹ Zhang et al. have fabricated carbon nanofibers/silver nanoparticles composite for catalytic reduction of 4-nitrophenol and found the maximum catalytic activity value of 6.2 s⁻¹ g⁻¹.⁷² The Ag nanoparticles-embedded mesoporous silica (AgNP@SiO₂) made by Pootawang et al. exhibited an activity parameter of only 0.36 s⁻¹ g⁻¹.⁷³

Table 3 Comparison of catalytic activities for the reduction of 4-NP with catalysts reported in the literature

Catalyst	k (s^−1)	k/M (s^−1 g^−1)	References
a The numbers in parentheses are obtained by considering the Ag NPs content alone.
Ag/C-0.05	5.32 × 10^−3	2.66	54
Ag-NP/C	1.69 × 10^−3	1.69	58
TAC-Ag-1	5.19 × 10^−3	1.3	68
Fe3O4@SiO2–Ag	7.67 × 10^−3	7.67	69
Ag-3/C	2.02 × 10^−2	4.03	18
CNFs/AgNPs	6.2 × 10^−3	6.2	70
AgNP@SiO2	3.6 × 10^−2	0.36	71
Fe3O4/SiO2–Ag	5.5 × 10^−3	275	72
Fe3O4@C@Ag–Au	15.80 × 10^−3	1580	17
AgNP-PG-5K	5.5 × 10^−3	1375	64
Ag(0.3)@SBA-15-(0)	3.4 × 10^−3	3.78 (1843)^a	This work
Ag(0.3)@SBA-15-(10)	1.1 × 10^−2	12.2 (4630)	This work
Ag(1.5)@SBA-15-(10)	2.3 × 10^−3	2.56 (809)	This work

---

Nevertheless, some literature has reported remarkably high activity parameters for catalytic reduction of 4-NP. For example, Shin et al. have found that the Fe₃O₄/SiO₂@Ag particles exhibited an activity parameter value of 275 s⁻¹ g⁻¹,⁷⁴ while An et al. reported Fe₃O₄@carbon microsphere supported Ag–Au bimetallic nanocrystals with an activity parameter value of 1580 s⁻¹ g⁻¹.²⁰ Baruah et al. have also demonstrated that the silver nanoparticles stabilized by cationic polynorbornenes can exhibit an activity parameter value of 1375 s⁻¹ g⁻¹.⁶⁶ However, these synthesis methods were somewhat tedious and time consuming. The activity parameter values of the present Ag(0.3)@SBA-15-(10) nanocatalyst are 12.2 and 4630 s⁻¹ g⁻¹, with and without consideration of the catalyst support, respectively, which are higher or comparable to the above mentioned Ag based catalysts. The high surface area of Ag(0.3)@SBA-15-(10) provides the active sites of Ag NPs to be easily accessible. Consequently, both 4-NP and BH₄⁻ can be efficiently adsorbed on the active sites of Ag(0.3)@SBA-15-(10), and thus the reduction reactions can be completed rapidly. Besides, our synthesis route is simple and straightforward in comparison to the above-mentioned complex systems. Therefore, the present Ag(0.3)@SBA-15-(10) nanocatalyst is a promising candidate for effective catalytic reduction of 4-nitrophenol.

To evaluate the catalyst efficiency, turnover frequency (TOF) of the catalysts was also calculated.⁷⁵ The TOF value for Ag(0.3)@SBA-15-(10) was found to be 1210 h⁻¹, which was higher than the Ag(0.3)@SBA-15-(0) value of 604 h⁻¹. Although higher TOF values indicated excellent catalytic activities for both Ag(0.3)@SBA-15-(10) and Ag(0.3)@SBA-15-(0), the former sample with –COOH functionality exhibited a much higher TOF value than the latter sample without –COOH functionality. This demonstrates that –COOH groups have inevitable roles in enhancing the catalytic activity by controlling the size of Ag NPs. The mono-dispersed and small sized Ag NPs inside the Ag(0.3)@SBA-15-(10) sample generates a larger number of active sites, which allows the effective contact between the reactants and the Ag NPs in the reaction and thus help to achieve a higher TOF value. On the other hand, the larger sized Ag NPs with slight particle aggregation on the Ag(0.3)@SBA-15-(0) sample (Fig. 3a) covered part of the active sites, leading to a lower TOF value. Another possibility is that the residual –COOH groups can probably transfer sufficient electron density to the surface of Ag NPs surface, which increases the activity of the Ag(0.3)@SBA-15-(10) sample and hence the TOF.¹⁷

One important factor to be considered for practical applications of catalysts is their degree of reusability and activity. The activity of the Ag(0.3)@SBA-15-(10) nanocatalyst was investigated in this study. When the first run of reaction was finished, the same catalyst was reused in successive reaction cycles by keeping all the reaction conditions identical. The kinetic constant of the first run is taken as a control, and the activity ratio is thus defined as the ratio of the kinetic constant in each run to the one of the first run. As shown in Fig. 7, the rate constants were slightly decreased in successive cycles. The phenomenon is expected since there may be some loss of Ag-NPs during the intermittent stage of sample washing.

Fig. 7 Activity of Ag(0.3)@SBA-15-(10) catalyst for the reduction of 4-NP with NaBH₄.

To explore the stability and leaching of Ag NPs after recycling tests, the recycled nanocatalyst was further characterized. As shown in Fig. 8A, the XRD pattern shows that the mesostructure is slightly deteriorated after successive cycling. The TEM image (Fig. 8B) shows that particle size of Ag NPs slightly increases from 3 nm to in the range of 4–6 nm. The increased particle size reduces the catalytic activity in the subsequent cycles as the surface area decreases. Still, mono-dispersed Ag NPs that are kept inside the mesopores of SBA-15 are observed after recycling tests. The nitrogen adsorption–desorption isotherm shown in Fig. 8C reveals that both surface area and pore volume are decreased, but pore diameter is increased after recycling tests. The repeated washing process after each cycle may etch the pore walls to certain extents, which enlarge the pore size and decrease the pore volume. The slight increase in pore size efficiently keeps the enlarged Ag NPs inside the pores and prevents a drastic change in the catalytic activity in recycling tests. The ICP-AES measurements were employed to explore the leaching level of Ag NPs after recycling tests. It was found that 16% of Ag NPs was leached out after the first cycle. The leaching level was slightly increased to 18% after the third cycle. Although significant leaching of Ag NPs (16%) was observed after the first cycle, there was only 2% leaching of Ag NPs from the first cycle to the third cycle, indicating that the nanocatalyst system was gradually stabilized. This demonstrates that the present Ag(0.3)@SBA-15-(10) catalyst system can be reused efficiently for several cycles without much Ag leaching.

Fig. 8 SAXS (A), TEM (B) and N₂ adsorption–desorption isotherm curve (C) of Ag(0.3)@SBA-15-(10) catalyst after 3 cycles.

Conclusions

In this study, the mesochannels of SBA-15 functionalized with carboxylic acid groups were demonstrated to be effective anchor sites for the growth of the Ag NPs with very small particle sizes down to 3 nm, which were well-dispersed within the mesopores without aggregation. Without –COOH moiety or too much silver contents, much larger Ag NPs with average diameters of ca. 20 nm were observed. The Ag NPs nanocatalysts thus obtained exhibited remarkable catalytic activity for the reduction of 4-NP, however, their activity strongly depended on their particle sizes. The larger the particle sizes, the lower the catalytic activity. In addition, these materials retained an almost complete conversion in the reduction reaction even they have been reused for five cycles. Thus, the current work presents a good example for using organic functionalized mesoporous silica materials as robust supports to fabricate metal nanoparticles within the mesopores. The same synthesis methodology can be readily applied to other noble metals and can effectively use as nanocatalysts for different catalytic reactions.

Acknowledgements

The financial support of this work by the Ministry of Science and Technology of Taiwan (Grant number: MOST 102-2113-M-008-006-MY3) is gratefully acknowledged.

Notes and references

1. M. Park, J. Im, M. Shin, Y. Min, J. Park, H. Cho, S. Park, M.-B. Shim, S. Jeon, D.-Y. Chung, J. Bae, J. Park, U. Jeong and K. Kim, Nat. Nanotechnol., 2012, 7, 803.
2. T. K. Sau, A. L. Rogach, F. Jäckel, T. A. Klar and J. Feldmann, Adv. Mater., 2010, 22, 1805.
3. S. Gao, N. Koshizaki, H. Tokuhisa, E. Koyama, T. Sasaki, J.-K. Kim, J. Ryu, D.-S. Kim and Y. Shimizu, Adv. Funct. Mater., 2010, 20, 78.
4. N. Yan, C. Xiao and Y. Kou, Coord. Chem. Rev., 2010, 254, 1179.
5. A. Roucoux, J. Schulz and H. Patin, Chem. Rev., 2002, 102, 3757.
6. J. D. Aiken III, Y. Lin and R. G. Finke, J. Mol. Catal. A: Chem., 1996, 114, 29.
7. Y. Lin and R. G. Finke, J. Am. Chem. Soc., 1994, 116, 8335.
8. C. Amiens, D. de Caro and B. Chaudret, J. Am. Chem. Soc., 1993, 115, 11638.
9. S. Chen and K. Kimura, J. Phys. Chem. B, 2001, 105, 5397.
10. A. Shahzad, W.-S. Kim and T. Yu, RSC Adv., 2015, 5, 28652.
11. M. Horecha, E. Kaul, A. Horechyy and M. Stamm, J. Mater. Chem. A, 2014, 2, 7431.
12. X. Liu, R. Jin, D. Chen, L. Chen, S. Xing, H. Xing, Y. Xing and Z. Su, J. Mater. Chem. A, 2015, 3, 4307.
13. S. M. El-Sheikh, A. A. Ismail and J. F. Al-Sharab, New J. Chem., 2013, 37, 2399.
14. J. Sun, D. Ma, H. Zhang, X. Liu, X. Han, X. Bao, G. Weinberg, N. Pfänder and D. Su, J. Am. Chem. Soc., 2006, 128, 15756.
15. Q. Geng and J. Du, RSC Adv., 2014, 4, 16425.
16. N. Yan, Y. Yuan and P. J. Dyson, Chem. Commun., 2011, 47, 2529.
17. Y. Yuan, N. Yan and P. J. Dyson, Inorg. Chem., 2011, 50, 11069.
18. N. Yan, Y. Yuan and P. J. Dyson, Dalton Trans., 2013, 42, 13294.
19. M. V. Patwadkar, C. S. Gopinath and M. V. Badiger, RSC Adv., 2015, 5, 7567.
20. Q. An, M. Yu, Y. Zhang, W. Ma, J. Guo and C. Wang, J. Phys. Chem. C, 2012, 116, 22432.
21. T. Ji, L. Chen, M. Schmitz, F. S. Bao and J. Zhu, Green Chem., 2015, 17, 2515.
22. Y. Li, Y. Cao, J. Xie, D. Jia, H. Qin and Z. Liang, Catal. Commun., 2015, 58, 21.
23. R. Rajesh, E. Sujanthi, S. Senthil Kumar and R. Venkatesan, Phys. Chem. Chem. Phys., 2015, 17, 11329.
24. J. Gao, J. Xu, S. Wen, J. Hu and H. Liu, Microporous Mesoporous Mater., 2015, 207, 149.
25. Z. Dong, X. Le, X. Li, W. Zhang, C. Dong and J. Ma, Appl. Catal., B, 2014, 158–159, 129.
26. J. Y. Ying, C. P. Mehnert and M. S. Wong, Angew. Chem., Int. Ed., 1999, 38, 56.
27. L. Fu, C. Huo, X. He and H. Yang, RSC Adv., 2015, 5, 20414.
28. Y. J. Han, J. M. Kim and G. D. Stucky, Chem. Mater., 2000, 12, 2068.
29. P. Wang, Z. Wang, J. Li and Y. Bai, Microporous Mesoporous Mater., 2008, 116, 400.
30. B. Naik, S. Hazra, V. S. Prasad and N. N. Ghosh, Catal. Commun., 2011, 12, 1104.
31. C.-M. Yang, P.-H. Liu, Y. F. Ho, C. Y. Chiu and K.-J. Chao, Chem. Mater., 2003, 15, 275.
32. S. S. Park, Y. K. Park, S. J. Choe and D. H. Park, J. Sol-Gel Sci. Technol., 2007, 42, 35.
33. A. Boullanger, S. Clément, V. Mendez, S. Daniele, C. Thieuleux and A. Mehdi, RSC Adv., 2013, 3, 725.
34. K. M. Ponvel, Y.-H. Kim and C.-H. Lee, Mater. Chem. Phys., 2010, 122, 397.
35. C.-S. Chen, C.-C. Chen, C.-T. Chen and H.-M. Kao, Chem. Commun., 2011, 47, 2288.
36. M. Merschdorf, W. Pfeiffer, A. Thon, S. Voll and G. Gerber, Appl. Phys. A, 2000, 71, 547.
37. C. T. Campbell, S. C. Parker and D. E. Starr, Science, 2002, 298, 811.
38. E. Katz and I. Willner, Angew. Chem., Int. Ed., 2004, 43, 6042.
39. Y. Li, Y. Wu and B. S. Ong, J. Am. Chem. Soc., 2005, 127, 3266.
40. R. R. Arvizo, S. Bhattacharyya, R. A. Kudgus, K. Giri, R. Bhattacharya and P. Mukherjee, Chem. Soc. Rev., 2012, 41, 2943.
41. J. Chen, X. Tang, J. Liu, E. Zhan, J. Li, X. Huang and W. Shen, Chem. Mater., 2007, 19, 4292.
42. W.-L. Dai, Y. Cao, L.-P. Ren, X.-L. Yang, J.-H. Xu, H.-X. Li, H.-Y. He and K.-N. Fan, J. Catal., 2004, 228, 80.
43. J. W. Medlin, J. R. Monnier and M. A. Barteau, J. Catal., 2001, 204, 71.
44. H. Liu, D. Ma, R. A. Blackley, W. Zhou and X. Bao, Chem. Commun., 2008, 2677.
45. K. Shimizu, K. Sawabe and A. Satsuma, Catal. Sci. Technol., 2011, 1, 331.
46. Y. C. Chang and D. H. Chen, J. Hazard. Mater., 2009, 165, 664.
47. J.-R. Chiou, B.-H. Lai, K.-C. Hsu and D.-H. Chen, J. Hazard. Mater., 2013, 248, 394.
48. H. Ozay, S. Kubilay, N. Aktas and N. Sahiner, Int. J. Polym. Mater., 2011, 60, 163.
49. M. Muniz-Miranda, Appl. Catal., B, 2014, 146, 147.
50. Y. Yang, Y. Guo, F. Liu, X. Yuan, Y. Guo, S. Zhang, W. Guo and M. Huo, Appl. Catal., B, 2013, 142–143, 828.
51. N. Sahiner, N. Karakoyun, D. Alpaslan and N. Aktas, Int. J. Polym. Mater. Polym. Biomater., 2013, 62, 590.
52. Y. Zhang, X. Yuan, Y. Wang and Y. Chen, J. Mater. Chem., 2012, 22, 7245.
53. S. Saha, A. Pal, S. Kundu, S. Basu and T. Pal, Langmuir, 2010, 26, 2885.
54. C.-T. Tsai, Y.-C. Pan, C.-C. Ting, S. Vetrivel, A. S. T. Chiang, G. T. K. Fey and H.-M. Kao, Chem. Commun., 2009, 5018.
55. M. Kruk and M. Jaroniec, Chem. Mater., 2001, 13, 3169.
56. D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548.
57. Y. Chi, J. Tu, M. Wang, X. Li and Z. Zhao, J. Colloid Interface Sci., 2014, 423, 54.
58. P. Scherrer, Göttinger Nachrichten Math. Phys., 1918, 2, 98.
59. S. Vetrivel and A. Pandurangan, J. Mol. Catal. A: Chem., 2004, 217, 165.
60. L.-Z. Wang, J.-L. Shi, J. Yu and D.-S. Yan, Nanostruct. Mater., 1998, 10, 1289.
61. S. Tang, S. Vongehr and X. Meng, J. Phys. Chem. C, 2010, 114, 977.
62. J. Liu, G. Qin, P. Raveendran and Y. Ikushima, Chem. –Eur. J., 2006, 12, 2131.
63. K. Mallick, M. Witcomb and M. Scurrell, Mater. Chem. Phys., 2006, 97, 283.
64. K. Mori, A. Kumami, M. Tomonari and H. Yamashita, J. Phys. Chem. C, 2009, 113, 16850.
65. J. Zhang, G. Chen, M. Chaker, F. Rosei and D. Ma, Appl. Catal., B, 2013, 132, 107.
66. B. Baruah, G. J. Gabriel, M. J. Akbashev and M. E. Booher, Langmuir, 2013, 29, 4225.
67. Z. Zhang, C. Shao, Y. Sun, J. Mu, M. Zhang, P. Zhang, Z. Guo, P. Liang, C. Wang and Y. Liu, J. Mater. Chem., 2012, 22, 1387.
68. J.-J. Lv, A.-J. Wang, X. Ma, R.-Y. Xiang, J.-R. Chen and J.-J. Feng, J. Mater. Chem. A, 2015, 3, 290.
69. W. Ye, J. Yu, Y. Zhou, D. Gao, D. Wang, C. Wang and D. Xu, Appl. Catal., B, 2016, 181, 371.
70. M. H. Rashid and T. K. Mandal, J. Phys. Chem. C, 2007, 111, 16750.
71. Y. Chi, Q. Yuan, Y. Li, J. Tu, L. Zhao, N. Li and X. Li, J. Colloid Interface Sci., 2012, 383, 96.
72. P. Zhang, C. Shao, Z. Zhang, M. Zhang, J. Mu, Z. Guo and Y. Liu, Nanoscale, 2011, 3, 3357.
73. P. Pootawang and S. Y. Lee, Mater. Lett., 2012, 80, 1.
74. K. S. Shin, Y. K. Cho, J.-Y. Choi and K. Kim, Appl. Catal., A, 2012, 413–414, 170.
75. L. Ma, L. Jia, X. Guo and L. Xiang, Chin. J. Catal., 2014, 35, 108.

---

Footnote

† Electronic supplementary information (ESI) available: EDS of Ag(0.3)@SBA-15-(10); FTIR spectra of Ag(y)@SBA-15-(x), where (y, x) = (a) (0, 10), (b) (0.3, 10), (c) (0.6, 10), (d) (0.9, 10), and (e) (1.5, 10); TGA curves of SBA-15-(10) and Ag(0.3)@SBA-15-(10); UV-Vis spectra of the successive reduction of 4-NP (a) before and (b) after addition of NaBH₄, and (c) SBA-15-(0). See DOI: 10.1039/c6ra01592a

---