Facile Fabrication and Raman Scattering Enhancement Properties of Mixed Gold and Silver Nanoparticle Layers

I have fabricated gold and silver nanoparticle layers on amino-terminated glass substrates by immersion processes from aqueous colloidal solutions of gold and silver nanoparticles. The composition ratio of nanoparticles on the modified substrates was varied by the mixing ratio of corresponding colloidal solutions of nanoparticles. Raman signals of rhodamine 6G were observed on the nanoparticle-modified substrates. In the case of the mixed gold and silver nanoparticle-modified substrates, the intensities of the Raman signals were larger than the intensities expected from a simple proportional sum of the Raman signals obtained using a gold nanoparticle layer or a silver nanoparticle layer. The nonlinear enhancement properties of the Raman signal intensities may be attributable to unique localized surface plasmon resonance between the gold and silver nanoparticles. [DOI: 10.1380/ejssnt.2012.157]

Controlling the near-surface electromagnetic field is an important factor for the application of plasmonic nanoparticles. For example, the enhancement factor of SERS is largely dependent on the size and shape of the nanoparticles. It is well known that strong Raman signals are observed at junctions of two adjacent particles, at protrusions of particles, and so on [1,[22][23][24]. At such a local nanospace, extremely strong LSPR is generated, which can also be estimated theoretically.
Combinations of different metal (ex. gold and silver) nanoparticles and core-shell architecture nanoparticles are quite interesting for developing plasmonic materials. Indeed, several groups reported that the considerably larger SERS was observed using bimetal plasmonic materials than that using corresponding monometal plasmonic material [25][26][27].
In this research, we have prepared mixed gold and silver nanoparticle layers on amino-terminated glass substrates by an immersion process, in which the mixing ratio of gold and silver nanoparticles was varied systematically. Raman scattering spectra of rhodamine 6G on the nanoparticle-modified substrates were measured and compared with each other.

II. EXPERIMENTAL
All chemicals were used as received. Water was used following distillation. Aqueous colloidal solutions of gold and silver nanoparticles (AuNPs and AgNPs) were prepared by the reduction of HAuCl 4 and AgNO 3 with trisodium citrate salt, which were based on the previous report [28]. Briefly, 190 mL of aqueous HAuCl 4 solution (2.5 × 10 −4 M) was refluxed. Next, 1.4 mL of aqueous trisodium citrate solution (1 wt%) was added to the refluxing solution, which was then refluxed for 1 h to obtain a colloidal AuNP solution. For the preparation of AgNPs, Extinction spectra of the aqueous colloidal solutions of nanoparticles and nanoparticle-modified glass slides were measured by a UV-vis-NIR spectrophotometer (V-670, JASCO). Raman scattering spectra of R6G-nanoparticlemodified substrates and the corresponding reference were obtained by a micro-Raman spectrometer (NRS-5100, JASCO). The excitation source was a 532 nm CW laser with a power of approximately 30 mW. The laser was focused onto the sample by using an objective lens (×100).
To reduce the degradation of the samples, the excitation laser was passed through a neutral density filter (OD= 3). Raman signals were averaged from 10 independent measurement points, which were obtained by one-shot measurements with 1 s of sampling time on each measurement point.

III. RESULTS AND DISCUSSIONS
Extinction spectra of aqueous colloidal solutions of AuNPs and AgNPs are showed in Fig. 2. A characteristic plasmon absorption band attributable to gold nanoparticles was observed around 530 nm in the solution of AuNPs. An obvious absorption peak around 410 nm attributable to a plasmon band of silver nanoparticles was also observed in the solution of AgNPs.
In AuNPs(0):AgNPs(100)/MAS, a similar pattern of absorption peaks was observed, with absorption bands around 400 and 670 nm. The former was attributable to isolated AgNPs, and the latter was attributable to interparticle plasmon coupling of aggregated AgNPs. For the mixed nanoparticle-modified substrates, the absorption profiles can nearly be interpreted as a mixture of AuNPs and AgNPs.
The extinction intensity at 532 nm for all nanoparticlemodified substrates is plotted in Fig. 4. The extinction intensity increased linearly with increasing ratio of Ag-NPs solution in the mixed AuNP:AgNP nanoparticle solution for preparation of nanoparticle modified substrates. Therefore, no obvious interaction between AuNPs and AgNPs was observed from the viewpoint of light absorption at 532 nm on the nanoparticle-modified substrate. On the other hand, the extinction intensity of R6G on AuNPs(n):AgNPs(m)/MAS at 532 nm was negligibly smaller than that of AuNPs(n):AgNPs(m)/MAS itself. Because, obvious extinction peak due to R6G was not observed in R6G/AuNPs(n):AgNPs(m)/MAS.  1128, 1184, 1310, 1361, 1509, 1573, and 1648 cm −1 from R6G were observed on the nanoparticle-modified substrates. In contrast, in the case of R6G spin-coated MAS glass slides (R6G/MAS), no obvious peak of Raman scattering from R6G was observed under same measurement condition. As described before, the extinction intensity of R6G on nanoparticle-modified substrates was almost negligible comparing with that of AuNPs and AgNPs. Therefore, AuNPs and AgNPs deposited on MAS-coated glass enhanced Raman scattering from R6G; qualitatively, Ag-NPs are more effective for enhancing Raman scattering than AuNPs. Figure 6 shows the relative Raman scattering spectra of samples after these spectral correc- These nonlinear enhancement properties of Raman scattering are quite interesting from a viewpoint of similarity with the enhancement of Raman scattering using combinations of gold and silver nanoparticles, core-shell nanoparticles and nano-sized hetero dimer [25][26][27].
In this research, detailed mechanism of the nonlinear enhancement of Raman scattering is not clear at stage. However, at least, the mixed layer of AuNPs and AgNPs may cause larger LSPR field between nanoparticles than that of monometal nanoparticle (AuNPs or AgNPs) layer. As a result, a nonlinear enhancement profile of Raman signals of R6G was seemingly obtained, which was depended on the mixing ratio of AuNPs and AgNPs solutions. After considering the effect of nanoparticle sizes and AuNPs-AgNPs ratio on glass substrates, a more detailed mechanism of enhanced Raman signals will be discussed. In addition, fluorescence emission enhancement of photoexcited dye molecules (e.g., porphyrins) using AuNP-AgNPmodified substrates will also provide valuable information about strong electric fields around nanoparticles or charge transfer between gold and silver nanoparticles. My current work focuses on this objective.

IV. CONCLUSION
I have succeeded in the fabrication of mixed gold and silver nanoparticles layers on amino-terminated glass substrates using immersion processes. Gold and silver nanoparticle layers with several mixing ratios of gold and silver nanoparticles were obtained from the mixed colloidal solutions of gold and silver nanoparticles. Raman signals from rhodamine 6G were strongly enhanced by the nanoparticle layers. The intensities of Raman signals of rhodamine 6G on mixed gold and silver nanoparticle layers were larger than that expected when gold or silver nanoparticle layers were used individually.