Abstract

Silver nanoparticles within the pore channels of selectively grafted mesoporous silica SBA-15 were synthesized. Silanols on the external surface of as-SBA-15 were first capped by – groups. After removal of the template of capped SBA-15 by calcination, silanols on the internal surface of SBA-15 were modified by 3-aminopropyltrimethoxysilane (APTMS), and then formaldehyde was grafted by amino groups of APTMS, and further Ag SBA-15). High-resolution transmission electron microscopy (HRTEM), X-ray diffraction (XRD), Fourier transformation infrared spectroscopy (FTIR), nitrogen adsorption/desorption isotherms, and UV-vis spectra confirm that the silver nanoparticles have been confined inside the channels of SBA-15. In addition, the SBA-15 modified electrode ( SBA-15/GC) exhibited an excellent electrocatalytic activity toward the reduction of hydrogen peroxide ( ). The proposed sensor exhibits a linear range of 48.5  M–0.97 M with a detection limit of 12  M ( ) and analytical time of 10 seconds per sample.

1. Introduction

Recently, the discovery of mesoporous silicas, such as M41s [1] and SBA-15 [2], has stimulated intensive studies of “host-guest” chemistry inside the channels of mesoporous silicas [36], which have potential applications in catalysis, selective adsorbents, medical [7], sensors [8, 9], and nanomaterials fabrications. Thanks to their uniform mesostructures, high surface areas, and tunable pore sizes [2], these ordered mesoporous silicas have been used as the promising templates to control the shape and size of metal nanoparticles [1016]. Many published works gave the relation between nanoparticles confined in various molecular sieves and their properties [1719]. Moreover, the surface of these ordered mesoporous silicas are also modified for many potential applications. For modifying the mesoporous materials through covalent linkage between functional groups and silica framework, two major methods, grafting (post synthesis) and cocondensation (direct incorporation), have been traditionally explored [20]. Grafting is one of the modification methods for presynthesized mesoporous silica, in which the organic functional groups can be introduced by direct reaction of organosilanes to silica surface. It is up to the application of thus formed materials, other functionalities can be fixed to the previously introduced functional groups including amino, thiol, and alkyl groups through covalent bonding and/or molecular recognition [2023]. The distribution and concentration of functional groups are influenced by reactivity of the organosilane and their accessibility to surface silanols, which are limited by diffusion and steric factors. Chao et al. [6, 11] prepared SBA-15 functionalized with (C O Si(C N(C Cl (TPTAC) and further synthesized metal nanoparticles by anion exchange between grafted SBA-15 and metal precursors inside the channels as well as upon reduction of precursors. The amount of metal loading as well as the morphology of metal in host SBA-15 can be rationally controlled through repeating ion-exchange/reduction cycles in the TPTAC-SBA-15 silica host. They used the same method to prepare Au nanoparticles and found that the size and morphology of Au nanoparticles in mesoporous SBA-15 are controllable by the preparation methods.

However, above-mentioned grafting methods basically allow introduction of functionalization at both intrapore and extrapore media, which led the nanoparticles form at both surfaces. Therefore, large metal particles aggregate would form on the external surface of the host materials. To overcome this disadvantage, Shi et al. [4] synthesized Pt nanoclusters within the pore channels of selectively modified mesoporous silica SBA-15 by a new in situ reduction process. The silanols on the external surface of SBA-15 were capped with –Si(C groups, thus effectively avoiding the formation of large particles outside the channels. On the other hand, the inner surface of the channel was functionalized with highly reducing Si–H bonds. Pt nanoclusters were formed inside the channels of SBA-15 from PtC by in situ reduction with Si–H bonds. Recently, Sun et al. [3] have developed a novel in situ autoreduction route to synthesize monodispersed silver nanoparticles inside the channels of SBA-15. It was demonstrated for the first time by the 13C CP/MAS NMR spectroscopy that amino groups of aminopropyltriethoxyl silane (APTS) modified mesoporous silica can be used to anchor formaldehyde to form NHC OH species, on which Ag(N N could be in situ reduced. The silver nanoparticles confined inside the channels of SBA-15 resulted in an unusual thermal stability.

Nanoparticles confined in SBA-15 silica possess high catalytic activity and stability thanks to the stabilizer-free and confined. Jiang et al. [24] found that the Pt nanoparticles confined in SBA-15 exhibit a high electrocatalytic activity toward the oxidation of carbon monoxide (CO) and methanol, and the linearly adsorbed CO species is the only intermediate derived from dissociative adsorption of methanol, which is more readily oxidized to form C in the aid of the active oxide in SBA-15.

Rapid, accurate, reliable, and reagentless determination of hydrogen peroxide ( ) is of great importance in food, clinical, pharmaceutical, industrial, and environmental analysis. Many analytical techniques have been reported for determination of hydrogen peroxide, for example, chemiluminescence [25], titrimetry [26], and spectrophotometry [25, 27]. Electrochemistry is an inexpensive and effective way to examine the reactions of many substances [27, 28]. Amperometric sensors are especially attractive because of their simplicity and high sensitivity. By now horseradish peroxidase (HRP), one of the most studied members of the family of heme enzymes, was shown to catalyze the reaction of nonmediated cathodic reduction of resulting from direct electron transfer from the electrode to the heme containing active site of HRP when immobilized at the electrode surface [29, 30]. Nevertheless, these systems were found to be too complex and the linear range for detection is narrow.

In this paper, the stabilizer-free and confined silver nanoparticles inside the channels of selectively grafted mesoporous silica SBA-15 by an in situ reduction process were synthesized (see Scheme 1). The catalytic activity of silver nanoparticles confined in the mesoporous silica SBA-15 to the reduction of hydrogen peroxide ( ) was studied for sensing application. The Ag-mSBA-15/GC modified electrode, thus allowed highly sensitive amperometric detection of , low applied potential, and a broad linear relationship with the concentration of over a wide range of  M. The new application of the stabilizer-free silver nanoparticles confined inside the channel of SBA-15 mesoporous silica, an attractive electrocatalytic nanomaterial for preparation of an amperometric sensor is proposed.

473791.sch.001

2. Experimental

2.1. Synthesis of Ag-mSBA-15

As illustrate by Scheme 1, mesoporous silica SBA-15 was synthesized following the published procedure [2] using the triblock copolymer Pluronic P123 as a template in acid conditions. Typically, a 6.4 g Pluronic P123 template was dissolved with stirring in a solution of 250 mL of 2 M HCl at 313 K, and 13.6 g of tetraethyl orthosilicate (TEOS) was then added. The resulting mixture was stirred at 313 K for 20 hours, and then aged at 373 K for 24 hours under static condition. The recovered solid was extensively washed with deionized water and drying at 353 K for 12 hours yielded as-SBA-15. To get calcinated SBA-15 (cal-SBA-15), the surfactant template of as-SBA-15 was removed by calcination in air at 823 K for 6 hours.

Ag-mSBA-15 was prepared according to the literature procedure [3]. 2.0 g as-SBA-15 was dispersed in 150 mL dry toluene under flowing , and then 10 mL trimethylchlorosilane (TMCS) was added dropwise under stirring. The mixture was filtered with toluene and ethanol after stirring continuously at 353 K for 8 hours. After that, the surfactant template was removed by calcination in air at 823 K for 6 hours. Thus, SBA-15 with the external surface capped with –Si(C (named cal-TMCS-SBA-15) was collected.

After having been outgassed at 353 K for 12 hours. 2.0 g cal-TMCS-SBA-15 was suspended in 150 mL dry toluene, and then 6.0 mL 3-aminopropyltrimethoxysilane (APTMS) was added under stirring. The mixture obtained was stirred for another 12 hours at room temperature and refluxed at 353 K for 8 hours. The solid was obtained after washing with toluene and then with ethanol intensively to eliminate the physically adsorbed APTMS and toluene. The selectively modified sample after being vacuum-dried at 353 K for 6 hours was labeled APTMS-TMCS-SBA-15.

To introduce reducer of formaldehyde into the channels, 1.0 g APTMS-TMCS-SBA-15 was soaked in a 105 mL mixture of formaldehyde, ethanol and water (formaldehyde/ethanol/water, 5 : 20 : 80, v/v/v), and the suspension was stirred at 313 K for 30 minutes. The product was filtered, rinsed with deionized water and dried at 323 K for 12 hours and denoted as HCHO-APTMS-TMCS-SBA-15.

For Ag incorporation, 1.0 g HCHO-APTMS-TMCS-SBA-15 was added into a mixture of ethanol and 0.01 M Ag(N N (aq) (1 : 4, v/v), and then the mixture was stirred at 313 K for 30 minutes. The product was filtered and rinsed thoroughly with deionized water and dried under vacuum at 323 K overnight, and thus formed confined silver nanoparticles were noted as Ag-mSBA-15.

In the control experiment, an aqueous solution of 1 mM NaB (70 mL) was cooled with ice and then 1 mM AgN aqueous solution (100 mL) was added to it under vigorous stirring, resulting in the light-brown Ag colloidal solution [31]. The product was filtered, rinsed with deionized water, dried and denoted as nm-Ag.

2.2. Characterization Methods

High-resolution transmission electron microscopy (HRTEM) images were obtained on instruments of FEI Tecnai-F30 electron microscopy operating at 300 KV. The powder samples were characterized by powder X-ray diffraction (XRD) using a Panalytical X'pert PRO diffractormeter (Tokyo, Japan) equipped with graphite monochromatized Cu Kalpha radiation. Surface functionalization was monitored by Fourier transformation infrared spectroscopy (FTIR) using a Nicolet FTIR 340 spectrometer. The pore diameter, pore volume, and surface area of the samples were derived from the nitrogen sorption isotherm at 77 K using a Micromeritics TriStar 3000 system. Prior to measurements, the sample was evacuated at 393 K for 5 hours. The UV-vis absorption was recorded on a Shimadzu UV-2100 spectrometer.

The material Ag-mSBA-15 was dispersed through ultrasonic vibration in a solution of dichlorethane containing polyvinyl chloride (PVC) to form a suspension. A defined quantity of the suspension was applied to a clean surface of glassy carbon (GC) substrate to form a thin film electrode, noted as Ag-mSBA-15/GC, then dried in the air for about 30 minutes. The solutions were prepared with Millipore water and chemicals of analytical grade. A saturated calomel electrode was used as reference electrode, and all electrochemical experiments were carried out at room temperature around 293 K. A CHI-660C potentiostat/galvanostat (Chenhua Instruments, Inc., Shanghai, China) was used in electrochemical studies.

3. Results and Discussion

3.1. Characterization
3.1.1. HRTEM Studies

In our synthesis procedures: (1) before removing the template, the external surface of the SBA-15 was firstly capped by trimethylchlorosilane (TMCS); (2) after removing the template, 3-aminopropyltrimethoxysilane (APTMS) molecules were introduced into the channels of SBA-15, which leads to that the amino functional groups were grafted to the internal walls of SBA-15 channels; (3) the reducers of formaldehyde were then introduced into the channels of SBA-15 and grafted on the amino groups inside the channel; (4) the metal precursors of Ag(NH3)2NO3 were finally introduced into the channels of SBA-15 by diffusion when the modified SBA-15 is put in a solution containing Ag(NH3)2NO3. The silver (0) nanoparticles were, therefore, formed through the reduction of Ag (I) ions by reducers previously fixed on the internal walls of the SBA-15 channels. Theoretically, Ag nanoparticles prepared by this method were located inside the channel between the walls in the method. The silver nanoparticles assembled inside the channels of SBA-15 can be further confirmed by HRTEM (Figure 1). Figure 1(a) shows the HRTEM of calcinated SBA-15 (cal-SBA-15), in which well-ordered channels are illustrated and are characteristic of mesoporous materials. Figure 1(b) depicts the HRTEM of Ag-mSBA-15, where the highly ordered pore structure of SBA-15 is still preserved. We can observe clearly that silver nanoparticles appear as dark spherical objects between the walls of SBA-15 and are homogeneously distributed inside the channels of SBA-15. It can be also seen that the size of Ag is slightly larger than the channel, which leads to the distortion of the channels.

3.1.2. XRD Studies

Figure 2(a) gives a small-angle XRD pattern of calcinated SBA-15(cal-SBA-15), three diffraction peaks appear in the spectrum, which are attributed to the characteristic diffraction peaks of (100), (110), and (200) for SBA-15, respectively, due to typical hexangular phase [2, 24]. After reaction with APTMS, the small-angle XRD pattern of APTMS-TMCS-SBA-15 (Figure 2(b)) still shows the characteristic diffraction peaks, which exhibits that the grafting of amino groups inside the channels did not affect the long-range ordering of the mesostructures. Compared with cal-SBA-15, the peaks shift to lower angles, and such shift may imply the enlargement of the frameworks [32]. The little negative shifting may be aroused by the covalent linkage between APTMS and hydroxy in the channels. We know from BET data in Table 1 the value is almost constant during the covalent linkage, but value has an obviously decrease after Ag nanoparticles were assembled. In addition, after the loading of Ag into APTMS-TMCS-SBA-15 (assigned as Ag-mSBA-15), the peaks (Figure 2(c)) have shifted positively comparing with that of APTMS-TMCS-SBA-15, the intensity of the peaks has decreased to a certain extent, while the full width at half-maximum (FWHM) of the peaks that is normalized by height has increased. These three features are attributed to contraction of framework during the support treatment, implying that Ag nanoparticles have been introduced into the channel of SBA-15 successfully as discussed elsewhere [24, 33]. The inset of Figure 2 is wide-angle XRD of Ag-mSBA-15. The broadening of the Ag diffraction peaks suggests that the size of Ag nanoparticles is in the nanometer range [8]. The average Ag particle size was estimated to be 8.0 nm, from the peak width of Ag (220) reflection by using Scherrer’s equation for approximation [34].

3.1.3. FTIR Studies

FTIR spectroscopy can provide surface information of materials for identification of chemical groups. Figure 3(a) presents the transmission FTIR spectrum of as-SBA-15, in which the bands in 2900–3000 cm−1 were attributed to template P123. After calcination in air at 823 K for 6 hours (cal-SBA-15), the bands in 2900–3000 cm−1 were disappeared (Figure 3(b)). This indicates that the template P123 was completely removed. After modification of the outer surface of as-SBA-15 with trimethylchlorosilane (TMCS), the sample was named as-TMCS-SBA-15. Figure 3(c) shows the FTIR spectrum of as-TMCS-SBA-15, in which the bands in 2900–3000 cm−1 attributed to template P123 and TMCS can be seen clearly. However, when the temple of the as-TMCS-SBA-15 has been removed by calcination in air at 823 K for 6 hours as it was done in Figure 2(b), we can still observe the bands in 2900–3000 cm−1 from Figure 3(d), which can be attributed to the C–H stretching modes of –C in cal-TMCS-SBA-15 [6]. This result confirmed that the methyl groups are still present on the material cal-TMCS-SBA-15 when template P123 was completely removed at 823 K. Such stability comes from the covalent linkage between –C groups and the hydroxy in the outer surface of silica framework.

The modification process of external surface of as-SBA-15 does not affect the internal surface of the channel because the template molecules occupy these channels during the external surface modification. The incorporation of APTMS inside the channels of the SBA-15 has been confirmed by transmission FTIR spectra as shown in Figure 4. The curve c is the spectrum of APTMS, and we can see the bands in the range 2900–3000 cm−1, which are attributed to the C–H stretching modes of –C in APTMS. The curve a is the spectrum of cal-TMCS-SBA-15, in which we can observe the characteristic absorption of Si–O–Si at 1082, 797, 465 cm−1 assigned to asymmetric stretching, symmetric stretching, and bend stretching [35]. In addition, the well-known IR adsorption bands which due to the stretching vibrational mode of surface silanol groups and water in the range 3500–3740 cm−1 [36, 37], and the C–H stretching modes of –C in the range 2900–3000 cm−1 [6] were also seen. The curve b showed the spectrum that APTMS was introduced into the channel of cal-TMCS-SBA-15, in which the intensity of IR absorption decrease for silanol group at 3500–3740 cm−1, and the intensity of IR absorption increase for the C–H group at 2900–3000 cm−1. The ratio of integrated intensity of the Si–OH band ( ) to integrated intensity of the C–H band ( ), that is, / , can be used to evaluate the increase of –C and –C and decrease of silanol due to the introduction of APTMS inside the channel of SBA-15. And the change of / can be used to monitor the reaction process between Si–OH and APTMS, since the intensity of the Si–OH band decreases and the intensity of the C–H band increases with the reaction progressing. Here, / is reduced from 38.1 at cal-TMCS-SBA-15 to 16.0 at APTMS-TMCS-SBA-15. The decrease of integrated intensity of silanol groups demonstrates an anchoring mechanism (see Scheme 1) involving reaction between Si–OH and APTMS [38].

3.1.4. Nitrogen Adsorption/Desorption Isotherms

Nitrogen adsorption/desorption isotherms for as-SBA-15, cal-SBA-15, cal-TMCS-SBA-15, APTMS-TMCS-SBA-15, HCHO-APTMS-TMCS-SBA-15, and Ag-mSBA-15 are found to be type IV isotherm curves with distinct hysteresis loops and steep adsorption/desorption steps were recorded that indicate a narrow pore size distribution. This suggests that the host silica was still maintained during the modification. Table 1 summarizes the results of desorption analyses. The specific surface area ( ) of as-SBA-15 is 262 m2/g. After silanols on the external surface of as-SBA-15 were capped by –Si(C groups, and the template was removed by calcination, increases to 705 m2/g. The total pore volume ( ), and the average pore size ( ) also show an increased value. After silanols on the internal surface of SBA-15 were grafted by 3-aminopropyltrimethoxysilane (APTMS), and of APTMS-TMCS-SBA-15 decrease from 705 to 419 m2/g and from 1.04 to 0.68 cm3/g, respectively, with a slight decrease of the from 6.57 to 6.49 nm. A considerable decrease in the , , and was measured in Ag-mSBA-15 due to the incorporation of Ag nanoparticles. The values of , , and for HCHO-APTMS-TMCS-SBA-15 are 335 m2/g, 0.48 cm3/g, and 5.86 nm, whereas they are 232 m2/g, 0.30 cm3/g, and 5.05 nm for Ag-mSBA-15. All these, together with the increased thickness of pore walls (t), can be attributed to the pore-filling effect [14, 39] and the fact demonstrated that the Ag nanoparticles have been confined inside the channel of SBA-15 [4].

The silver nanoparticles were synthesized in the inner surface of silica framework by the “ship-in-a-bottle” approach, in which the channel of SBA-15 is used as a microreactor to prepare the nanoparticles. Precursors were introduced into the channels one by one, and were assembled into Ag nanoparticles in the microreactor. This resulted in the Ag nanoparticles were confined inside the channel of SBA-15. Silanols on the external surface of as-SBA-15 were first capped by –Si(C groups. After removal of the template of capped SBA-15 by calcination, silanols on the internal surface of SBA-15 were modified by 3-aminopropyltrimethoxysilane (APTMS), and then formaldehyde was grafted by amino groups of APTMS to form NHC OH species, on which Ag(N N could be in situ reduced into Ag nanoparticles. Without silanols on the surface of SBA-15, there will be impossible to form NHC OH species, and further to produce Ag nanoparticles. So there will be no Ag nanoparticles on the external surface of SBA-15 thanks to silanols on the external surface first capped by –Si(C groups to decrease its activity. of APTMS-TMCS-SBA-15 is 6.49 nm from BET data, and the average Ag nanoparticle size was estimated to be 8.0 nm from XRD analysis. HRTEM image has also demonstrated that the size of Ag is slightly larger than the channel, which leads to the distortion of the channels. Therefore, we infer that the channels are slightly distorted during the nanoparticles synthesized in the microreactor, which makes the size of nanoparticles prepared inside the channel of SBA-15 larger than the size of the channel diameter. As a result, the problem of catalyst leaking is much lessened or eliminated. Similar results have been also reported in literature [40].

3.1.5. UV-vis Absorption Spectroscopic Studies

UV-vis absorption spectra of the Ag-mSBA-15 samples may be used to provide additional evidence of the formation of Ag nanostructures inside the channels of SBA-15 powder and reveal unique optical properties. Figure 5(a) (line 1) shows the UV-vis absorption spectrum of HCHO-APTMS-TMCS-SBA-15, in which a spectrum with a nearly linear was recorded. Figure 5(a) (line 2) shows the absorption spectrum of Ag-mSBA-15, in which only the peak of 420 nm which assigned to the surface plasmon resonance (SPR) of Ag nanoparticles [16] is observed and the peak of 345 nm assigned to small silver clusters from 2 to 8 atoms formed in micropores of silica walls [41] is absent. SPR is a characteristic feature of metal nanoparticles between the sizes of 2 and 50 nm [42]. This result indicates that the reduction of silver (I) ions in the micropores of the SBA-15 does not exist, since all the reducers were introduced into the mesopores of the SBA-15 in our experiment.

Moreover, silver nanoparticles have been studied as a catalyst in reduction reactions of dyes like methylene blue (MB) [43]. Here, reduction of MB by NaB is used as a standard for determining the catalytic activity of Ag-mSBA-15. The preliminary catalytic testing for Ag-mSBA-15 was carried out by reduction of MB (2 mL  M) in water using Ag-mSBA-15 (1 mg) as catalyst and NaB (1 mL  M) as the reducing agent. The progression of the catalytic reduction of MB can be easily followed by the change of absorbance intensity at 665 nm that is absorbance maximum ( ) of MB. The dot line in Figure 4(b) is the UV-vis spectrum of the mixture containing a mixture of MB dye plus NaB reducing agent, with pure silica SBA-15, which gives a strong peak at 665 nm attributing to of MB, illustrating that MB have not been reduced. When 1 mg Ag-mSBA-15 was added into above solution, it can be seen the band at 420 nm attributing to the SPR of Ag nanoparticles Figure 5(b) (solid line). The absorbance at of MB gradually decreases with the reaction time, which suggests MB begin to reduce, and meanwhile we can observe the color of solution changes from blue to colorless. Above investigations have illustrated that silver nanoparticles exhibit a high catalytic activity to the reduction of MB.

3.2. Application of Ag-mSBA-15 as a Sensor for Detecting Hydrogen Peroxide
3.2.1. Electrochemical Response of Ag-mSBA-15/GC Electrode to Reduction of

Figure 6(a) displayed the cyclic voltammetry (CV) of glassy carbon (GC) in 0.2 M HAc-NaAc buffer solution (pH 5) containing 3 mM , in which a featureless CV was observed, and the current may be mainly ascribed to the double-layer charging of the electrode. Figure 6(b) is CV of Ag-mSBA-15/GC in 0.2 M HAc-NaAc buffer solution (pH 5), in which a small reduction current appears that may be attributing to reduction of support electrolyte. Figures 6(c), 6(d), and 6(e) show the CVs of Ag-mSBA-15/GC, nm-Ag/GC, and bulk Ag, respectively, in 0.2 M HAc-NaAc + 3 mM . It can be seen a wave at −0.42 V for Ag-mSBA-15/GC, −0.65 V for nm-Ag/GC, and −0.53 V for bulk Ag appear. Ag-mSBA-15/GC, nm-Ag/GC, and bulk Ag electrode on the initial reduction potential of is −0.12 V, −0.15 V, and −0.17 V, respectively. The reduction potential of on Ag-mSBA-15/GC was 140 mV, 230 mV, and 110 mV more negative than that on silver nanoparticles assembles supported on GC [44], nm-Ag/GC, and bulk Ag, indicating a lower overpotential for reduction at the Ag-mSBA-15/GC. Figure 7 depicts the CVs of Ag-mSBA-15/GC in 0.2 M HAc-NaAc + 3 mM with different scan rate. It can be seen that the peak currents of the reduction increase in Ag-mSBA-15/GC in 0.2 M HAc-NaAc + 3 mM with an increasing scan rate. In the scan rate range from 10 mV·s−1 to 500 mV·s−1, the reduction current of is proportional to the square root of scan rate (inset of Figure 7), which suggests that the rate of electrochemical reaction is rather fast and the electrode process is controlled by the diffusion of from solution to electrode surface. In order to determine the optimal working potential for the sensing, the electrochemical response of the was researched in different potentials. The relationship between the steady-state current and the operating potential in 0.2 M HAc-NaAc + 3 mM at Ag-mSBA-15/GC is showed in Figure 8. Considering the economy of energy, the sensitivity and the steadiness of Ag-mSBA-15/GC, −0.45 V were chosen as optimal working potential for sensing with amperometry of constant potential.

3.2.2. Detection of

Figure 9(a) shows the dynamic response of the Ag-mSBA-15/GC modified electrode at a working potential of −0.45 V with successive injections of 3 mM in 0.2 M HAc-NaAc buffer solution (pH 5). The calibration curve of Ag-mSBA-15/GC modified electrode under the optimal working potential of −0.45 V is showed in Figure 9(b). Under an exactly same condition, the Ag-mSBA-15/GC modified electrode for determination exhibits a linear range of  M with a detection limit of  M (S/N = 3), and analytical time of 10 seconds per sample. The relative standard deviation (RSD) is 2.8 % for ten repetitive measurements of 3 mM solution. While the amperometric response of the bulk Ag electrode to shows a narrower linear relation in the range of  M with a detection limit of  M, and analytical time of 10 seconds per sample.

The reproducibility and storage stability of the sensor were examined. When Ag-mSBA-15/GC modified electrode was stored in air and subjected to the day-by-day calibrations at room temperature, the electrode can maintain over 97% of the initial value in the response to 3.0 mM after 100 days, while the bulk Ag electrode decayed quickly to 92% after 24 hours. Since the size of silver nanoparticles synthesized inside the channel of SBA-15 is larger than the channel diameter, the problem of catalyst leaking is much lessened or eliminated. In addition, silver nanoparticles were confined inside the channel of SBA-15, which made the catalyst stable and anticontaminated.

As a comparison, we have modified also the synthesis procedures of Ag-mSBA-15 simply skipping the step TMCS treatment. In this case, both inside and outside of the SBA-15 channels were grafted by reducers, and Ag (0) nanoparticles were produced both inside and outside of the SBA-15 channels. We have tested the stability of this sample; and the results demonstrated that its activity for reduction was declined at the first 15 days, then stabilized for long time as the sample of Ag-mSBA-15. The results confirmed that the Ag (0) nanoparticles inside the SBA-15 channels are stable, while those outside the SBA-15 channels are less stable.

In the experimental, it may be always difficult to attain an entire consistent with the theoretical design of preparation of the catalyst. So, a small portion of sites in external surface of the SBA-15 may not have been capped by trimethylchlorosilane (TMCS), and lead to form some Ag (0) nanoparticles outside the SBA-15. However, as reported in UV-vis spectra, we have not observed the peak at 345 nm in UV-vis spectra of samples, which indicated that the quantity of clusters from 2 to 8 atoms could be neglected. In addition, if a small quantity of Ag (0) nanoparticles were formed outside the SBA-15 channels, they are not stable. As a consequence, they will not affect the stability of the sensor that uses the Ag (0) nanoparticles catalysts confined in the SBA-15 channels.

4. Conclusions

The new application of the stabilizer-free silver nanoparticles confined inside the channel of SBA-15 mesoporous silica (Ag-mSBA-15) is proposed. The uniform mesostructures, high surface areas, and tunable pore sizes of SBA-15 facilitates its manipulation for sensor preparation and sensing application. The resulting Ag-mSBA-15 modified electrode shows a very efficient electrocatalytic behavior toward the reduction of at a low overpotential. The sensor for exhibits very good analytical performance with low cost, convenient preparation, and sensitive and rapid detection. Thus, the Ag-mSBA-15/GC is an attractive amperometric sensor for and other practical applications.

Acknowledgment

The study was supported by grants from Natural Science Foundation of China (20433040, 20573085).