Silver nanoparticles deposited on pyrogenic silica solids: Preparation and textural properties

Introduction

Silver nanoparticles can be supported on different substrates and matrices for obtaining materials with desirable properties (Isse et al., 2006; Jiang and Liu, 2003; Neelgund and Oki, 2011; Nguyen et al., 2013). Among many possibilities, hydrophilic silicon dioxide products are applied as potential support for noble metal ions, nanoparticles, and many other types of active phases and components (Zebardastan et al., 2016). The producers show many advantages of pyrogenic silicon dioxide such as high chemical purity, good insulation properties, particle nature, and assurance as to the reproducibility. In the potential applications, it is worthwhile emphasizing its controlled surface chemistry (being a result of continuous production process, adjustable surface area and textural properties), easy-to-use, reliability, and efficiency. Nanocomposites based on silicon dioxide and nanoscale noble metal particles show high performance with combination special properties from both components. The immobilization of noble metal nanoparticles onto the inorganic supports allows to control the morphology and behavior of nanomaterials as well as their stabilization by avoiding coalescence and aggregation due to the eliminating van der Waals forces and high surface energy. In the case of silver-based nanocomposites, many studies are associated with obtaining homogeneous coatings using various methods of their preparation (Chen and Zhang, 2003; Shen et al., 2014). As examples, we may suggest preparation of new type of substrate for surface-enhanced Raman scattering (Alak, 1989; Willetta and Chumanov, 2016), materials for catalytic reduction of dyes (Jiang et al., 2005), catalysts for CO oxidation (Qu et al., 2015) or elements of advanced membrane devices (Huang et al., 2014). In this work, we propose the preparation of silver-based nanocomposites by combining both high chemical reactivity of nanosilver phase and material with defined surface chemistry. Despite the tremendous progress in the field of preparation and application of new nanocomposites and silver-based materials, it is still desirable to develop a simple, low-cost, and environmentally friendly approach for preparing SiO₂/Ag composites.

Experimental

Pyrogenic silica under the trade name of AEROSIL®300 (Asil 300) were purchased from Evonik Degussa GmbH, Germany. 3-Aminopropyltriethoxysilane (APTES), silver nitrate, and toluene were purchased from Sigma-Aldrich.

Chemical modification of Asil300 surface was performed using 3-APTES as grafted molecules to obtain support with high affinity for silver ions and ability to immobilize the silver nanoparticles. Preparation of amino-functionalized materials was performed by addition of 3.0 g of the dried Asil300 to 50 mL of toluene and appropriate amount of APTES for creation half and full monolayer of amino groups (0.3 and 0.6 mmol/g, respectively) based on primary silanol groups on silica surface. The resultant mixtures were stirred and refluxed at 150℃ for 2 h. Amino-functionalized silicon dioxide particles was decorated with silver ions by wet impregnation approach. Final silver nanoparticles were formed by thermal reduction of silver ions at 300℃. The diamminesilver(I) complex [Ag(NH₃)₂]⁺ was applied as a silver precursor and prepared according to description specified in our earlier article (Zienkiewicz-Strzalka et al., 2016). Textural properties of initial and modified materials were obtained by measuring N₂ adsorption/desorption isotherms at −196℃ using a Micromeritics ASAP2020 equipment. Specific surface areas were estimated by applying the Brunauer–Emmett–Teller (BET) theory. Data analysis was performed using MicroActive software (Micromeritics). Mesopore pore-size distribution curves were obtained from the desorption branch of isotherm using the Barrett–Joyner–Halenda (BJH) model with cylindrical pores and Faas correction without smooth differentials. Pore size distribution (PSD) functions of pores in the micropore range were calculated using Horvath–Kawazoe (HK) method (cylinder geometry (Saito-Foley)) and non-local density functional theory (NLDFT) method (cylinder pore geometry) using non-negative regularization of 0.001. Transmission electron microscopy (TEM) measurements were carried out with a high resolution scanning transmission electron microscope Titan G2 60-300 (FEI).

Results and discussion

Nitrogen physisorption isotherms were used to estimate the specific surface area and related textural properties of investigated materials before and after modification by silver nanophase. The changes of textural properties of the pristine porous structure and samples after amino-functionalization and further silver nanoparticles deposition were observed. The experimental nitrogen adsorption–desorption isotherms and PSDs of investigated samples are presented in Figure 1. The shape of all isotherms (reversible type II) seems to correspond to the model of multilayer adsorption in low-porous or macroporous materials. In the relative pressure range above p/p₀ = 0.8, the capillary forces are responsible for filling of large mesopores and macropores. This phenomenon is illustrated by quite narrow hysteresis loop in this area of isotherm. The changes of the adsorbed amounts of nitrogen were observed for investigated samples. Quantity of adsorbed nitrogen decreases with increasing of amino group and silver nanoparticle content. The textural properties of investigated samples are collected in Table 1. The specific surface area (S_BET) of the initial pyrogenic silica sample was determined as 265 m²/g with pore volume determined as single point adsorption total pore volume at p/p₀ = 0.98 as 1.14 cm³/g. The specific surface area of micropores was determined as 59 m²/g, i.e. 22% of the total surface area.

OPEN IN VIEWER

Figure 1.

(a) Experimental nitrogen adsorption–desorption isotherms at 77 K for Asil300 silicon dioxide samples before and after deposition of silver nanoparticles and the textural properties of investigated samples as an inset (S_Total: surface area calculated using experimental points at relative pressure of (P/P₀) 0.035–0.31; V_Total: pore volume calculated by 0.0015468 amount of nitrogen adsorbed at P/P₀ = 0.99; S_mic, V_mic: surface area and pore volume of micropores calculated by t-plot method with fitted statistical thickness in the range of 4.5 to 6.0 Å; D_h: hydraulic pore diameters calculated according equation: Dh = 4V_Total/S_Total; D_mo(NLDFT), D_mo(HK): the pore diameters estimated from PSD maximum of NLDFT and HK, respectively). (b) Effect of the sample functionalization on the characteristics of t-plot curves. (c,d) Evolution of the pore size distribution obtained from NLDFT using model of cylindrical pores for oxide surface (Tarazona, 1985; Jaroniec et al., 2000; Tarazona et al., 1987). (e) Mesopore size distribution obtained from BJH model.

OPEN IN VIEWER

Table 1.

Textural properties of investigated materials.

Sample	Surface area (SBET) (m2/g)		Pore volume (cm3/g)		Pore size distribution (nm)
	STotal	Smic	VTotal	Vmic	Dh (4V/S)	Dmo (NLDFT)	Dmo (HK)
Asil300	262	59	1.14	0.03	17.4	2.1	2.0
Asil300_NH2_03	206	14	2.38	0.008	46.2	2.6/6.5	2.4
Asil300_NH2_06	149	13	1.46	0.005	39.1	3.0	2.5
Asil300_Ag	132	30	0.57	0.015	17.2	2.3	2.3
Asil300_NH2_03_Ag	104	5	1.17	0.001	35.4	2.9	2.4
Asil300_NH2_03_Ag	65	8	0.66	0.004	40.6	3.0	2.6

After functionalization by 3-APTES, the surface area decreases with increasing the amount of incorporated functional groups. Deposition of the silver nanoparticles generates additional significant changes of the porous structure and further reduction of the available surface. The proportional reduction of specific surface after silver nanoparticle deposition was observed according to the trend discussed above.This suggests that voids of materials are occupied by silver nanoparticles. Figure 1(b) shows t-plot curves for investigated nanocomposites and fitted lines (gray area from t = 4.5 to 6 Å indicates fitted points). The Y-intercept of the lines corresponds to the volume of micropores and shows lower values for 3-APTES modified samples and even greater decrease of this parameter for samples modified by silver nanoparticles. In this case, the larger silver nanoparticles can block entry to the micropores. In this work, the BJH and NLDFT methods were used to determine the PSDs from nitrogen sorption isotherms. In Figure 1(c) and (d), PSD curves illustrate the presence significant volume of micropores (average diameter ∼2 nm). Due to the surface modification, the amount of micropores as well as some shift in their average size was observed. In all the cases, modification (both by amino groups as well as silver nanoparticles) causes the decrease in amount of micropores. Pyrogenic silica based nanocomposites also exhibit significant amount of macropores, which were achieved after modification (size above 100 nm) (Figure 1(e)). In the case of Asil300_NH₂_06 sample, the calculated PSDs suggest shift of maximum pore size to 3 nm. At the same time, the amount of macropores decreased significantly. Such differences between samples may suggest that when certain amount of 3-APTES is applied (0.6 mmol/g), the micropores (range of pore sizes below 3 nm) and macropores are first occupied by modifier.

To perform measurement of the particle size, both primary particles of silicon dioxide and silver nanostructures, TEM was applied. Figure 2 shows spherical structure of the support and well-defined deposited silver nanocrystalites.

OPEN IN VIEWER

Figure 2.

(a,b) TEM images of pyrogenic silicon dioxide aggregates modified by silver nanoparticles. (c,d) TEM images with high magnification (HRTEM) of silver nanoparticles deposited on silica surface.

The TEM images show that the visible aggregates consist of primary silica particles formed during flame process. The silica particles create merging zones with amorphous structure where SiO₂ tetrahedrons exhibit an irregular arrangement. This organization of tetrahedrons is faintly visible in the amorphous area of composites (inset of Figure 2(c)). Silver nanoparticles are visible as dark objects on the pyrogenic silica surface (Figure 2(b)). The distribution of silver phase is uniform but the size of silver crystallites can be differentiated depending on the investigated area. Silver phase exists mostly as small crystallites with size below 5 nm. The aggregation effect has not been observed to a large extent. HRTEM images confirmed that diamminesilver(I) complex as silver precursor in combination with hydrophilic silica ensure the creation of composites where silver phase forms well-defined crystallites with well-singled lattice planes without clear marked internal distortion (Figure 2(c) and (d)).

Thermal stability of investigated samples was evaluated by thermogravimetric (TG) analysis before and after silver nanoparticles incorporation. The weight loss starting around 300℃ and is related to decomposition of amine groups on the surface (4.2% for sample where initial content of amino groups was 0.6 mmol/g and 2.56% for sample with 0.3 mmol/g of amino groups). The total weight loss at 950℃ was estimated as 9% and 5.5%, respectively. The presence of silver ions had no significant effect on thermal stability of the composites.

Conclusion

In this work, we investigate the properties of silver nanoparticles supported on hydrophilic pyrogenic silicon dioxide materials functionalized by amino groups. Silver nanoparticles deposited on silica surface create homogeneous phase without aggregation to dysfunctional component. Such behavior of silver phase can be related to the presence of 3-ammoniatriethoxysilane groups, which can be responsible for homogeneous distribution due to their affinity to noble metal nanoclusters. The presence of silver nanoclusters on the silica surface generates changes over the textural properties reducing the specific surface area and available adsorption sites, however does not block them entirely making the material still useful as a potential sorption agent.