Microbead silica decorated with polyhedral silver nanoparticles as a versatile component of sacrificial gel films for SERS applications

Abstract

A new method of microbead silica preparation with superficially built-in 2–3 nm silver seeds is suggested by using a simple reaction of Stoeber SiO₂ microspheres with hot aqueous solutions of diamminesilver(I) hydroxide without addition of reducing agents. The seeding beads initiate growth of polyhedral 20–50 nm silver nanoparticles encrusting silica surface after an instant heterogeneous contact with a mixture of silver nitrate and ascorbic acid solutions with concentrations of as low as 1 mM. The unique microstructure results in a pronounced, about 100 nm, red shift of the silver plasmonic band allowing fast and robust tuning of optical properties of the nanocomposites. Such microengineered building blocks are stored safely in a sacrificial biopolymer (ethylcellulose) film as its versatile component to be applied on demand in aqueous environment since polymer swelling absorbs water-soluble analytes and allows for the advanced SERS analysis.

---

Introduction

Silver nanocomposite materials attract permanent interest in various applications due to their unique properties as catalysts or building blocks of complex materials with strong plasmon resonance for advanced SERS applications in biology, medicine and ecology.^1–12 This is fortunately combined with relatively high stability and facile fabrication methods; seeded growth and other techniques are designed to obtain anisotropic silver nanoparticles or those of a complex morphology such as nanoflowers, nanorices, cubs, multipods and nanodendrites, mesocages.^10–36 Various strategies are developed for the solution-based synthesis of metal nanocrystals with branched morphologies including kinetically controlled overgrowth, aggregation-based growth, heterogeneous seeded growth, template-directed methods; selective, anisotropic etching and colloidal system ageing.^1,2,10,12 From a chemical point of view, most of the proposed methods are based on a balance between nucleation, growth and facet blocking processes and that all results in an advanced control over optical properties of the nanoparticles for their possible SERS functions.

Another effective strategy to design SERS-active materials is based on preparation of nanocomposites of microparticles like Stoeber-derived silica microbeads³⁷ decorated with nanoparticles.^38–45 Such materials allow to concentrate nanoparticles on the surface of dielectric microcarriers, to organize additional hot spots, to build up self-assembling structures and to control optical properties by thickness and coverage of the SiO₂ core by the silver shell.^38–45 Usually, silver reduction agents like ascorbic acid, NaBH₄, Sn(II), polyvinylpyrrolidone (PVP) are used to produce and electrostatically attract silver nanoparticles to the surface of silica^39–45 while it is shown in a number of recent papers that diamminesilver(I) complexes can be used alone to produce silver nanoparticles under mild conditions.^38,46–50

In this paper, a new method of microbead silica preparation with superficially incorporated silver seeds is suggested by using a facile reaction of Stoeber SiO₂ microspheres with hot aqueous solutions of diamminesilver(I) hydroxide without an addition of reducing agents as followed by further overgrowth of polyhedral silver nanoparticles resulting in a pronounced and controllable shift of the silver plasmonic band. That allows fast and robust tuning of optical properties of the nanocomposites remaining biocompatible due to the absence of toxic reagents in their processing. Finally, we suggested to apply the materials as a SERS-active part within a protecting, sacrificial biopolymer to be applied on demand in aqueous environment causing polymer swelling and absorption of water-soluble analytes.

Experimental

Preparation of silver decorated silica beads included two basic steps-seeding and subsequent overgrowth of silver nanoparticles onto SiO₂ microspheres. All the reagents were of high quality and purchased from Aldrich; high purity water (Milli-Q, Millipore) was used to prepare all the solutions and for washing of all the products. Silica microspheres were obtained in a standard way³⁷ using the ammonia-catalyzed hydrolysis of tetraethyl orthosilicate (TEOS) at 40 °C for 5–10 h with molar ratios of components as C₂H₅OH : NH₃ : TEOS : H₂O = 63 : 13 : 4 : 1. The as-prepared suspension was repeatedly centrifugated (6000 rpm, 15 min) and washed (redispersed) with water to remove residual reagents.

To prepare aqueous solutions of diamminesilver(I) hydroxide, 0.1 M aqueous sodium hydroxide was added dropwise to a freshly prepared 0.01 M aqueous silver nitrate solution until complete precipitation of a black–brown silver(I) oxide. This as-prepared oxide was thoroughly washed with water and dissolved in a two-fold molar excess of a 10% aqueous ammonia solution (prepared from 30% ammonium hydroxide) to yield 0.001–0.05 M solution of a silver(I) complex ([Ag(NH₃)₂]OH). The resulting transparent precursor was filtered through Millex-LCR syringe driven filter units (Millipore, 0.45 μm pores).

At the seed preparation stage, silver complex concentration (0.001–0.05 M), silica to the silver complex ratio (100 : 1–1 : 1) and soaking time (10–100 min) were varied. In some experiments (ESI, Fig. S1†), silver–silica nanocomposites were treated repeatedly with the hot complex solution, or a mixture of an aqueous silver complex mixed with silica suspension was injected into boiling water (ESI, Fig. S2†). In a typical experiment with optimized conditions, 0.3 g of Stoeber silica beads dried at room temperature were placed in a 50 ml glass with 0.01 M of [Ag(NH₃)₂]OH + 10 mol% of excessive NH₃ and heated up rapidly to 90 °C, magnetically stirred for a half of hour, then the yellowish suspension was cooled down rapidly, centrifuged, treated with 10 wt% ammonia to remove possible traces of silver(I) oxide. Usually, the colour of the solid phase did not change after that and XRD confirmed the presence of metallic silver (ESI, Fig. S3†). Then the silver–silica seeds were redispersed in pure water and centrifuged twice. This procedure, in principle, allows to produce easily a large amount of seeding silver–silica suspension. We tested also the effectiveness of small amounts of additives of possible reducing agents like ethanol or PVP (see below) but they do not show much better results compared to pure silver complex. The amount of silver content in the seeding microspheres was estimated as about 6% with respect to Si by EDX (ESI, Table S1†). The advantages of this particular seeding procedure over traditional approaches (ESI, Fig. S4 and S5†) include its easiness, uniformity of silver seeds distribution and their narrow particle size of about 2–5 nm.

At the silver overgrowth stage, the seeding silver–silica microspheres were placed into silver nitrate water solutions of the 0.1–0.001 M concentration range followed by an addition of equal volume aliquots of ascorbic acid of the same concentration as silver nitrate. The color of mixtures has stopped to change finally for about 10–20 seconds thus confirming the completion of the redox reaction. Then, centrifuging, redispersing and washing procedures were repeated at least twice to ensure the storage of the silver–silica nanocomposite in pure water. The weight ratio for those suspensions was set as 100 mg of the nanocomposite per 5 ml of water. The overgrowth procedure can be repeated several times to increase coverage of the silica beads by silver and to change the size/shape of the encrusted nanoparticles.

The sacrificial biopolymer films were prepared by addition of 10–20 vol% of 0.5–2 wt% solution of ethylcellulose to the above described suspensions of the nanocomposites followed by tape casting and room temperature drying on a usual aluminum foil; the 0.5–2 mm thick layers of the liquid were used to produced polymer films of a different thickness and of variable silver–silica concentrations (ESI, Fig. S6–S9†). It was easy then to defoliate thicker films to get them free-standing. Either free-standing or aluminum-attached films were cut off into about 5 × 5 sq. mm pieces to perform SERS measurement by addition of a single droplet of a water soluble analyte and its soaking by the swelling polymer film (ESI, Fig. S10†) containing the above described SERS-active silica–silver nanocomposite.

The obtained silver nanostructures were studied by a transmission electron microscopy (TEM) and electron diffraction (ED) using LEO912 AB OMEGA (Carl Zeiss) and Jeol JEM 2100F. The sample microstructures were studied additionally using a field emission high resolution scanning electron microscope (Carl Zeiss NVision 40) at 0.5–5 kV accelerating voltage equipped with an EDX detector (X-Max, Oxford instruments). All the size distribution and zeta potential measurements were performed using a Malvern Zetasizer Nano ZS instrument at 25 °C with a He–Ne laser (laser power 4 mW, wavelength 632.8 nm and beam diameter 0.63 nm). The scattering and detection angles were 175 and 12.8°, respectively. The samples were placed in standard Malvern zeta potential disposable capillary cells and polystyrene cuvettes for zeta potential and size measurements, respectively. All the measurements were repeated several times. UV-vis absorption spectra were recorded using the UV-vis spectrophotometer Lambda 35 (Perkin-Elmer) in a diffuse reflectance mode. The phase composition was examined using Rigaku D/MAX 2500 (Japan) with a rotating copper anode (CuK_α irradiation, 5–90° 2θ range, 0.02° step). Diffraction maxima were indexed using the PDF-2 database. Raman spectroscopy and SERS experiments were performed using InVia Raman microscope (Renishaw, UK) equipped with 514 and 633 nm lasers and power neutral density filters (0.05–10%). All the spectra were collected using ×20 objective lens and 10–180 s of acquisition time (draft and final spectra). To calibrate the system, a silicon wafer was applied. Macroscopic images were taken using Tamron SP 60 mm f2 macro lenses attached to a Canon EOS 650D digital camera in a manual mode.

Results and discussion

Traditional generation strategies of heterogeneously seeding SiO₂ microspheres include either separate preparation of seeds by homogeneous nucleation followed by their electrostatic attachment to silica^39–42 or forming special superficial groups or chemisorbed strong reducing agents to provide predominant formation of silver seeds directly onto the surface of silica^40,43,45 (as also shown in ESI, Fig. S4 and S5†). We consider another, much simpler and effective, way supported by our previous works^46–50 which is based on gradual self-reduction of one of the most useful and well-known ammonia complex of silver, [Ag(NH₃)₂]⁺, due the lost of its stability upon the ammonia ligand leakage (ESI, Fig. S1†).

This route utilizes drastic stabilization of silver ions within the complex (K_stab([Ag(NH₃)₂]⁺) = 1.3 × 10⁷, Ag⁺ + 2NH₃ = [Ag(NH₃)₂]⁺), huge variation of the redox potential of silver in the presence of ammonia, a possibility to precipitate the Ag₂O solid phase or dissolve it by ammonia (Ag₂O + H₂O + 4NH₃ = 2[Ag(NH₃)₂]⁺ + 2OH⁻) and, finally, the absence of side-anions in the redox system except OH⁻, which is a natural part of all the aqueous solutions (K_w = [H⁺][OH⁻] = 10⁻¹⁴). Another part of the system is oxygen which can be dissolved in the solution as an oxidizing agent or evolved from the solution as a product of silver oxide decomposition, the standard potential of the pair O₂/OH⁻ is . That is why the overall reaction (1)

Ag + 1/4O₂ + 2NH₃ = [Ag(NH₃)₂]⁺ + OH⁻ (1)

serves as a good control tool for generation of silver nanoparticles simply by changing the concentration of ammonia which could generate enough Ag⁺ ions to cause their reduction to metallic silver otherwise Ag⁺ is stabilized as [Ag(NH₃)₂]⁺ and can not be reduced.

The electrode potentials for the redox pairs are the follows:

O₂ + 2H₂O + 4e = 4OH⁻, E_O2/OH⁻ = 0.401 + 1/4 × 0.059 log(P_O2/[OH⁻]⁴) (2)

Ag = Ag⁺ + e, E_Ag/Ag⁺ = 0.8 + 0.059 log[Ag⁺] (3)

The overall redox potential E_react1 for the reaction (1) is a combination of the reactions (2) and (3), E_O2/OH⁻–E_Ag/Ag⁺, or, for P_O2 = 0.21 atm, E_react1 = (−0.399 − 0.01γ) − 0.059γ log([OH⁻][Ag⁺]) where γ = T/298 gives the potential change with temperature. “Free” [Ag⁺] is expressed by the ratio [Ag(NH₃)₂⁺]/(K_stab[NH₃]²) while electroneutrality conditions dictate that [OH⁻] = [Ag(NH₃)₂⁺] + [Ag⁺] + (1.8 × 10⁻⁵[NH₃])^1/2. The latter comes from OH⁻ generation in ammonia solutions, NH₃ + H₂O = NH₄⁺ + OH⁻ (K_amm = 1.8 × 10⁻⁵) although [Ag(NH₃)₂⁺] dominates since NH₃ is a weak base and contributes negligibly.

The calculated data of the E_react1 potential are presented in Fig. 1. It is evident that even very small concentrations of ammonia makes the reaction (1) to shift to the right side forming [Ag(NH₃)₂]⁺, no silver nanoparticles could be formed, they will be dissolved due to stabilization of silver ions within the complex. However if the NH₃ ligand is lost for some reasons (heating and evolving from the solution), there is an area of negative potentials where diamminesilver(I) hydroxide is not stable thermodynamically and decomposes forming metallic silver in the reversed reaction (1) which deposits as nanoparticles. The ABC pathway means a starting point of a high concentration of separately prepared [Ag(NH₃)₂]⁺ in an excess of ammonia, as described in the experimental part, then a sharp drop of the concentration of ammonia should happen, for example, due to the ligand evaporation with partial decomposition of the complex, and then the process ends up at a very small concentration of [Ag(NH₃)₂]⁺ and a higher concentration of “naked” silver ions.

Fig. 1 Calculated potential for the reaction Ag + 1/4O₂ + 2NH₃ = [Ag(NH₃)₂]⁺ + OH⁻ as a function of ammonia concentration in boiling water, the lower curve is equal to 0.1 M concentrations of [Ag(NH₃)₂⁺], the upper one is diluted, 10⁻⁴ M, solution of the complex. The ABC direction corresponds schematically to the formation of silver in experiments.

The experimental results on the ABC reaction pathway (Fig. 1) are shown in Fig. 2 as accomplished after boiling the silver complex together with silica beads. The silver–silica nanocomposite forms after the interaction of hot silver complex with silica beads as seen from TEM images, SAED and XRD data (Fig. 2, see also ESI, Fig. S2 and S3†). A typical size of the isotropic silver nanoparticles embedded (encrusted) into the superficial layer of silica is surprisingly small, only about 2–5 nm, and demonstrates no large deviation in the mean size (Fig. 2b). We observed similar small “self-reduced” nanoparticles from diamminesilver(I) complex earlier^46,47 and the data are also confirmed elsewhere.³⁸ It seems that such a small size is typical for such transformation of the silver complex, moreover it depends weakly on treatment time or complex concentration (ESI, Table S1†). The observed seeding nanoparticles are spherical while longer time or repeated treatment with a hot silver complex could cause formation of new, larger, overgrown nanoparticles of silver (Fig. 2a) of about 20–25 nanometers in diameter and those particles are twinned (Fig. 2c). In all the above given cases, no polygonal, polyhedral or distinctly facetted particles are found.

Fig. 2 Typical TEM images of silica beads after treating with a hot solution of [Ag(NH₃)₂]OH. (a) Two fractions of silver–silica nanocomposites, 1 – major fraction of Stoeber silica microspheres with incorporated 2–5 nm silver seeds (1), and a minor fraction with both seeds and as-grown 15–20 nm silver nanoparticles (2), the arrow indicates silica surface etched by the base solution of [Ag(NH₃)₂]OH with increased OH⁻ concentration compared to aqueous NH₃. The upper inset shows SAED data for the major fraction, the lower inset gives the overall XRD pattern of the dried sample, the arrow shows the presence of amorphous silica, (b) the beads with seeds and (c) silica microspheres with enlarged silver nanoparticles, magnified views with notations: 1 – silica microsphere body, 2 – silver seeds or nanoparticles.

Another feature of the silica beads treated in a hot solution of the diamminesilver(I) complex is superficial etching (Fig. 1a, see also ESI†). That is caused chemically by partial silica dissolution due to a reaction with hydroxide ions present in the aqueous silver complex as given by formal eqn (4) and (5):

[Ag(NH₃)₂]OH = [Ag(NH₃)₂]⁺ + OH⁻ (4)

SiO₂ + 2OH⁻ = SiO₃²⁻ + H₂O (5)

It seems that this etching process is important for further overgrowth of silver since many originally formed seeding nanoparticles go deeply into the body of silica microspheres, they are encrusted into the superficial layer but not simply attach to the surface, as typical for traditional methods (ESI, Fig. S4 and S5†). Unfortunately, as known, 2–5 nm silver nanoparticles are not optimal for SERS^1,2,5 and thus a stage of overgrowth is crucial.

The results of the second, overgrowth, stage are shown in Fig. 3 and 4. The ascorbic acid is a strong reducing agent with respect to silver ions and thus mixing of silver nitrate solutions with solutions of ascorbic acid leads normally to immediate deposition of black metallic silver (Fig. 3a, “L8”–“L10”). This occurs due to homogeneous nucleation of silver after a redox reaction and fast growth of silver particles above 100 nm (Fig. 3g). The latter is not plausible since silver forms everywhere in the bulk solution while coverage of silica by silver remains insufficient. Moreover, such a large particle size also gives, as well-known, materials with small SERS effectiveness (Fig. 4d, “1”, see also, for example, papers^2,5).

Fig. 3 Microstructural features of silver–silica composites. (a) A macroscopic view of aqueous sols, seeding samples: s1 – boiling of Stoeber silica in 0.01 M [Ag(NH₃)₂]OH for 30 min, s2 – same with an addition of 0.5 vol% of ethanol, s3 – same as “s1” with an addition of 0.5 vol% of 2 wt% solution of polyvinylpyrrolidone, grown samples g4, g5 and g6 – “s1”, “s2” and “s3” sols, respectively, after 5–10 s of addition of a 1 : 1 mixture of 1 mM AgNO₃ and 1 mM of ascorbic acid, r7 – the colourless “redox” nutrient 1 : 1 mixture itself of 1 mM AgNO₃ and 1 mM of ascorbic acid; black suspensions of L8, L9, L10 are similar to g4, g5 and g6 but concentrations of AgNO₃ and ascorbic acid are set 10 times higher, 0.01 M; (b and c) – SEM view of the sample “g4” in secondary electron and back – scattered electron modes, respectively, (d) – a magnified SEM view of the sample “g4” showing silica beads (1), pits (2) on the surface of silica and polygonal silver nanoparticles (3). (e–g) – SEM images of the samples “g5”, “g6” and “L8”, respectively, 1 – silica microspheres, 2 – large grains of silver. In the latter case, volumetric formation of reduced metallic silver (2) occurs contrary to superficial growth of silver nanoparticles onto silica in the previous samples. The top panel demonstrates a scheme of sample preparation and particle size distribution for samples selected for SERS measurements (“g4” mainly).

Fig. 4 Sacrificial polymer films containing silica beads with polygonal silver nanoparticles. (a) Bubbled coatings on an aluminum foil (left) and a corresponding free-standing film (right, light transmission) of suspension of the sample shown in Fig. 3d tape-casted with an addition of 1 wt% of ethylcellulose and then dried at room temperature; (b) concentration of ethylcellulose is reduced down to 0.5 wt%; (c) an optical image on a mirror showing changes in color in a transmission and reflective configurations; (d) diffuse scattering spectra of the dried samples shown in Fig. 3a, 1 – suspension “L8”, 2 – “g5”, 3 – “g6”, 4 – “g4”, respectively.

A decrease in the concentration of the solutions for overgrowth by a couple of orders of magnitude prevents homogeneous nucleation (Fig. 3a, “L7”) but remains effective for seeded growth of silver nanoparticles directly on the surface of silica beads (Fig. 3a, “g4”–“g6”, 3b–f). In our experiments, such a drop in the amounts of reducing and oxidizing agents generating threshold supersaturation with respect to silver was performed by mixing solutions of AgNO₃ and the ascorbic acid of millimolar concentrations. According to the Nernst equation and related Gibbs energy changes, such dilution decreases much the driving force of silver precipitation and also hinders the kinetics of the process. As a result, a homogeneous nucleation barrier can not be overcome and only heterogeneous nucleation proceeds giving decorated silver–silica nanocomposites (Fig. 3). Interestingly, small amounts of additives of possible reducing agents like ethanol or PVP (Fig. 3a, “s2”, “s3”, “g5”, “g6”, 3e and f) do not show much better results compared to pure silver complex (Fig. 3a, “s1”, 3b–d).

The most intriguing feature of the samples prepared in the present work is the unexpected formation of unusual facetted/polyhedral silver nanoparticles on the surface of silica beads (Fig. 3d). Gentle centrifugation of the coloured suspensions (those in Fig. 3a, “s1”–“g6”) deposits the silica nanocomposites completely while the supernatant remains colourless and transparent with no extra nanoparticles observed. Therefore overgrowth in diluted solutions of silver nitrate and ascorbic acid provides predominantly heterogeneous nucleation followed by anisotropic particle formation without application of usual surfactants or citrate ions known to produces such nanoparticles due to selective grain growth.^2,5,27 The particles in Fig. 3c and d are not electrostatically attached to the surface, as usual, but encrusted due to overgrowth from the seeding particles resided in small concaved pits (see, for example, Fig. 3d, “2”) formed by the surface etching with alkaline ammonia complex of silver during the seeding stage (Fig. 2). Experimentally, the nanocomposite particles do not change the silver coverage after ultrasonication for 30 minutes and only rare growth pits inside SiO₂ would be seen in micrographs (Fig. 3d, “2”). It should be noted that the shape of many particles in Fig. 3d is not cubic as typical of the equilibrium habitus of FCC phases but triangles, angled triangles grow in the nanocomposite samples from the surface of silica and hexagonal platelets are also clearly seen being a quite unusual shape of facetted silver nanoparticles^52,53 especially keeping in mind free growth of the nanoparticles without usual surfactants like PVP or citrate ions.

Such a nanocomposite reveals evidently a pronounced red shift of the plasmonic band, actually till the beginning of the common light absorption interval of spherical gold nanoparticles (∼520 nm) (Fig. 4d), and this is observed visually as an unusual deep purple colour of thin concentrated layers of the nanocomposite within a xerogel of ethylcellulose (Fig. 4a and c), the dichroism typical of anisotropic nanoparticle ensembles⁵¹ is also evident from Fig. 4c as a greenish colour on reflection (see also ESI, Fig. S6–S7†).

It is known that for an FCC metal like silver, (111) faces are more stable than (100) ones. The growth rates of the two faces are much more different in the presence of PVP or citrates causing selective grain growth. For example, citrate might bind more strongly to the (111) side planes than the (100) side plane, resulting that the relative growth rate of the (100) is significantly larger than that of the (111) side planes. Thus, the (100) faces vanish and larger triangular nanoparticles with (111) side planes form. As usually observed, nanoprisms are enlarged, the areas of the (111) side planes increase and more citrate molecules are needed to stabilize the crystal surface. If the concentration of citrate in solution is negligible, the inhibition of citrate on the nanoparticle will be absent therefore silver will tend to deposit onto the (111) side planes. Hence, the difference between growth rates of (111) and (100) should not be significant and the nanoparticles will change from triangular nanoplates to hexagonal nanoplates.

Thus, most probably^52,53 that the observed hexagonal nanoplates have an FCC structure (XRD shows this phase, Fig. 1) and the (111) planes are the basal planes of the nanoplates. The two lateral faces are consistent with (100) and (111) faces. It is possible that the nanoplates appear as flat crystals with two (111) faces at the top and the bottom, limited by three straight (111) faces at the edges and by three (100) faces at the corners. Hence, their outside shape appears as hexagonal. This is achieved by a simple variation of supersaturation in the redox reaction with the ascorbic acid but not by the selective grain growth if surfactants are applied.

The resulting nanocomposites are reasonable to store inside a sacrificial biopolymer film like ethylcellulose (Fig. 4) which is a good candidate as a carrier of nanoparticles⁵⁴ that could prevent environmental damage of the nanoparticles and would also make it suitable to apply the material on demand for an advanced SERS analysis. Fig. 5, a demonstrates a SERS signal enhancement for small pieces of cellulose free-standing films encapsulating a small portion (a droplet) of the rhodamine 6G model analyte in concentration as low as 10⁻⁸ M due to polymer swelling (ESI, Fig. S10†). We used a very low laser intensity (0.5–1% of maximal power) to provide very soft, noninvasive conditions for the analysis. Interestingly that the 514 nm laser irradiation allows also to detect medically important analytes like bilirubin at low concentrations (Fig. 5a). In these cases, especially for bilirubin, the plasmon band position of the Ag@SiO₂ nanocomposite (Fig. 4d) favors the observed SERS enhancement of the spectral signal.

Fig. 5 Application of sacrificial polymer films containing silica beads with polygonal silver nanoparticles (“g4” sample in Fig. 3d) for SERS studies using different laser irradiation: (a) 514 nm wavelength (0.5% non-damaging laser power, 120 s of acquisition time, the inset shows SERS spectra for 1% of power and 20 s signal accumulation from the films), 1, 3, 4–10⁻⁸ M dye rhodamine 6G, 2–10⁻⁷ M bilirubin solution in toluene, the peaks are assigned by ref. 55, (b) 633 nm wavelength (10% laser power, 10 s of acquisition time), 1 – Raman spectra of blank dried 0.5 wt% ethylcellulose film on aluminum foil, the star marks the strongest Raman peak from ethylcellulose, 2 – Raman signals from concentrated Ag@SiO₂ nanocomposite without ethylcellulose and with no dye molecules, 3 – reference Raman spectrum of methylene blue dye with 10⁻⁷ M and 10⁻⁴ M (“4”) concentrations in pure ethylcellulose film on aluminum, 5 – SERS spectrum of 10⁻¹⁰ M methylene blue in sacrificial polymer films with about 10 wt% of Ag@SiO₂.

At the same time, a long tail of light absorption (Fig. 4d) allows to apply another laser irradiation, 633 nm, to agitate Raman signals (Fig. 5b). As evident, a cellulose film itself demonstrates a few weak peaks only in the region of 1050–1100 cm⁻¹ overlapping with no important peaks of the discussing analytes (marked with a star in Fig. 5b, “1”). The dry Ag@SiO₂ nanocomposite itself shows also several small peaks (Fig. 5b, “2”) since raw SiO₂ microbeads could contain residual substances remaining after TEOS hydrolysis. However “dilution” of the nanocomposite by mixing with ethylcellulose, forming then a xerogel film, results in a great enhancement of the methylene blue signal with concentration as low as 10⁻¹⁰ M (Fig. 5b, “5”). The ordinary Raman signal of this dye becomes detectable at the level of about 10⁻³ to 10⁻⁴ M only without Ag@SiO₂ (Fig. 5b, “3” and “4”) making an enhancement factor to exceed 10⁶ in this case. Thus, all the samples are SERS-active and exhibit a new example of robust and flexible technology to produce SERS materials with adjustable functional properties.

Conclusions

A new method of microbead silica preparation with superficially incorporated almost monodispersed 2–3 nm silver seeds is suggested by using a simple reaction of Stoeber SiO₂ microspheres with hot aqueous solutions of diamminesilver(I) hydroxide without addition of reducing agents. The seeding beads become decorated with polyhedral 20–50 nm silver nanoparticles after an instant heterogeneous contact with a mixture of silver nitrate and ascorbic acid solutions with concentrations of as low as 1 mM otherwise large silver particles are formed homogeneously. The unique microstructure results in a pronounced, about 100 nm, red shift of the silver plasmonic band allowing fast and robust tuning of optical properties of the nanocomposites remaining biocompatible due to the absence of any toxic reagent in their processing. It was suggested that such building blocks could be stored safely in a sacrificial biopolymer (ethylcellulose) film as its versatile component to be applied on demand in any aqueous environment since polymer swelling absorbs water-soluble analytes and allows for the advanced SERS analysis. The whole process becomes possible due to an interplay of redox potentials of pure silver(I) cations and its ammonia complex that results in the observed morphology. The method suggested would be a new facile and flexible experimental procedure to form silver nanocomposites for practical SERS applications.

Acknowledgements

Authors thank Russian Science Foundation for financial support (grant 14-13-00871). The authors acknowledge partial support from M. V. Lomonosov Moscow State University Program of Development for providing analytical research of the materials.

Notes and references

1. L. Polavarapu, J. Perez-Juste, Q. Xu and L. M. Liz-Marzan, J. Mater. Chem. C, 2014, 2, 7460–7476.
2. R. A. Alvarez-Puebla and L. M. Liz-Marzan, Energy Environ. Sci., 2010, 3, 1011–1017.
3. D. Cialla, A. Maerz, R. Boehme, F. Theil, K. Weber, M. Schmitt and J. Popp, Anal. Bioanal. Chem., 2012, 403(1), 27–54.
4. M. Moskovits, Rev. Mod. Phys., 1985, 57, 783–826.
5. D. D. Evanoff Jr. and G. Chumanov, ChemPhysChem, 2005, 6, 1221–1231.
6. B. Lim and Y. Xia, Angew. Chem., Int. Ed., 2011, 50(1), 76–85.
7. X. K. Meng, S. C. Tang and S. Vongehr, J. Mater. Sci. Technol., 2010, 26(6), 487–522.
8. B. Wiley, Y. Sun, B. Mayers and Y. Xia, Chem.–Eur. J., 2005, 11(2), 454–463.
9. Q. Zhang, Y. N. Tan, J. Xie and J. Y. Lee, Plasmonics, 2009, 4(1), 9–22.
10. R. A. Alvarez-Puebla and L. M. Liz-Marzán, Small, 2010, 6(5), 604–610.
11. K. Chaloupka, Y. Malam and A. M. Seifalian, Trends Biotechnol., 2010, 28, 580–588.
12. Y. D. Tretyakov and E. A. Goodilin, Russ. Chem. Rev., 2009, 78(9), 801–820.
13. J. Fang, S. Liu and Z. Li, Biomaterials, 2011, 32, 4877–4884.
14. J. K. Daniels and G. Chumanov, J. Phys. Chem. B, 2005, 109(38), 17936–17942.
15. A. März, B. Mönch, P. Rösch, M. Kiehntopf, T. Henkel and J. Popp, Anal. Bioanal. Chem., 2011, 400(9), 2755–2761.
16. T. Huang and X.-H. N. Xu, J. Mater. Chem., 2010, 20(44), 9867–9876.
17. X.-Y. Zhang, A. Hu, T. Zhang, W. Lei, X.-J. Xue, Y. Zhou and W. W. Duley, ACS Nano, 2011, 5(11), 9082–9092.
18. S. Peng and Y. Sun, Chem. Mater., 2010, 22, 6272–6279.
19. B. Pietrobon and V. Kitaev, Chem. Mater., 2008, 20, 5186–5190.
20. N. Leopold and B. Lendl, J. Phys. Chem. B, 2003, 107(24), 5723–5727.
21. S. C. Boca, M. Potara, A.-M. Gabudean, A. Juhem, P. L. Baldeck and S. Astilean, Cancer Lett., 2011, 311(2), 131–140.
22. D. E. Charles, D. Aherne, M. Gara, D. M. Ledwith, Y. K. Gun'ko, J. M. Kelly, W. J. Blau and M. E. Brennan-Fournet, ACS Nano, 2010, 4, 55–64.
23. D. D. Evanoff Jr. and G. Chumanov, J. Phys. Chem. B, 2004, 108, 13948–13956.
24. D. Yu and V. W.-W. Yam, J. Phys. Chem. B, 2005, 109, 5497–5503.
25. J. Zeng, X. Xia, M. Rycenga, P. Henneghan, Q. Li and Y. Xia, Angew. Chem., Int. Ed., 2011, 50(1), 244–249.
26. L. Fan and R. Guo, Cryst. Growth Des., 2008, 8(7), 2150–2156.
27. A. Guerrero-Martínez, S. Barbosa, I. Pastoriza-Santos and L. M. Liz-Marzán, Curr. Opin. Colloid Interface Sci., 2011, 16, 118–127.
28. B. K. Jena, B. K. Mishra and S. Bohidar, J. Phys. Chem. C, 2009, 113(33), 14753–14758.
29. H. Liang, Z. Li, W. Wang, Y. Wu and H. Xu, Adv. Mater., 2009, 21, 4614–4618.
30. X. S. Shen, G. Z. Wang, X. Hong and W. Zhu, Phys. Chem. Chem. Phys., 2009, 11(34), 7450–7454.
31. J. Zhang, H. Liu, P. Zhan, Z. Wang and N. Ming, Adv. Funct. Mater., 2007, 17, 1558–1566.
32. E. C. Cho, C. M. Cobley, M. Rycenga and Y. Xia, J. Mater. Chem., 2009, 19, 6317–6320.
33. Y. Sun and Y. Xia, Science, 2002, 298, 2176–2179.
34. S. Coskun, B. Aksoy and H. E. Unalan, Cryst. Growth Des., 2011, 11(11), 4963–4969.
35. D. Chen, X. Qiao, X. Qiu, J. Chen and R. Jiang, Nanotechnology, 2010, 21, 025607.
36. H. Liang, H. Yang, W. Wang, J. Li and H. Xu, J. Am. Chem. Soc., 2009, 131, 6068–6069.
37. W. Stöber, A. Fink and E. Bohn, J. Colloid Interface Sci., 1968, 26, 62–69.
38. X.-H. Liu, Y.-Y. Cao, H.-Y. Peng, H.-S. Qian, X.-Z. Yang and H.-B. Zhang, CrystEngComm, 2014, 16, 2365–2370.
39. Z. Deng, M. Chen and L. Wu, J. Phys. Chem. C, 2007, 111, 11692–11698.
40. D. Radziuk, R. Schuetz, A. Masicb and H. Moehwald, Phys. Chem. Chem. Phys., 2014, 16, 24621–24634.
41. M. Jasiorski, K. Łuszczyk and A. Baszczuk, J. Alloys Compd., 2014, 588, 70–74.
42. J. C. Flores, V. Torres, M. Popa, D. Crespo and J. M. Calderyn-Moreno, J. Non-Cryst. Solids, 2008, 354, 5435–5439.
43. M. Zhu, G. Qian, Z. Wang and M. Wang, Mater. Chem. Phys., 2006, 100, 333–336.
44. K. Wang, X. Zhang, C. Niu and Y. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 1272–1278.
45. T. Liua, D. Li, D. Yanga and M. Jiang, Colloids Surf., A, 2011, 387, 17–22.
46. A. A. Semenova, N. A. Brazhe, E. Y. Parshina, V. K. Ivanov, G. V. Maksimov and E. A. Goodilin, Plasmonics, 2014, 9(2), 227–235.
47. A. A. Semenova, E. A. Goodilin, A. N. Brazhe, V. K. Ivanov, A. E. Baranchikov, V. A. Lebedev, A. E. Goldt, O. V. Sosnovtseva, S. V. Savilov, A. V. Egorov, A. R. Brazhe, E. Y. Parshina, O. G. Luneva, G. V. Maksimov and Y. D. Tretyakov, J. Mater. Chem., 2012, 22(47), 24530–24544.
48. A. A. Semenova, V. K. Ivanov, S. V. Savilov and E. A. Goodilin, CrystEngComm, 2013, 15(39), 7863–7871.
49. A. S. Sarycheva, A. A. Semenova, E. Y. Parshina, N. A. Brazhe, A. Y. Polyakov, A. Y. Kozmenkova, A. V. Grigorieva, G. V. Maksimov and E. A. Goodilin, Mater. Lett., 2014, 121, 66–69.
50. A. V. Sidorov, O. E. Vashkinskaya, A. V. Grigorieva, T. N. Shekhovtsova, I. A. Veselova and E. A. Goodilin, Chem. Commun., 2014, 50, 6468–6470.
51. N. G. Khlebtsov, Anisotropic properties of plasmonic nanoparticles: depolarized light scattering, dichroism, and birefringence, J. Nanophotonics, 2010, 4(1), 041587.
52. H. Jia, J. Zeng, J. An, W. Song, W. Xu and B. Zhao, Thin Solid Films, 2008, 516, 5004–5009.
53. X. Zou, E. Ying, H. Chen and S. Dong, Colloids Surf., A, 2007, 303, 226–234.
54. F. Jiang and Y.-L. Hsieh, Biomacromolecules, 2014, 15, 3608–3616.
55. B. Yang, R. C. Taylor and M. D. Morris, Spectrochim. Acta, Part A, 1993, 49, 1735–1749.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental and characterization details. See DOI: 10.1039/c5ra16788d

---