The transmission electron microscopy (TEM) images of the nanoparticles were obtained with a Libra-120 microscope (Carl Zeiss, Oberkochen, Germany). The zetapotential of the particles was measured before and after drying with a Zetasizer Nano-ZS instrument (Malvern Instruments, Malvern, UK). The silica spheres were fabricated by the Stöber method [54] by adding the desired amount (from 0.1 to 1 mL) of 25% aqua ammonia to 10 mL of absolute ethanol and then magnetically stirring (500 rpm) the solution obtained for 5 min at room temperature. Thereafter, 0.3 mL of tetraethyl orthosilicate

was added dropwise, and the suspension was stirred for 1 h and then left to stay overnight without stirring. The size of the silica spheres (200 nm in our case) is governed by the amount of ammonia added. The fabricated silica

spheres were deposited by spin coating at 2,000 rpm on silicon wafers by means of a homemade centrifuge and then heat-treated [55]. The substrates BMN 673 cost were examined by scanning electron microscopy (SEM) using a JSM-6700 F instrument (JEOL, Akishima-shi, Japan), atomic force microscopy (AFM), and absorption spectroscopy with a Shimadzu UV-3600 UV–vis spectrophotometer (Shimadzu Corporation, Kyoto, Japan). The AFM images were obtained with an INTEGRA-Therma AFM microscope (NT-MDT, Moscow, Russia) operated in the semicontact and phase-contrast modes. The overall resolution was 512 × 512 points for a 2 × 2 μm2 region. The SERS selleck spectra were measured with an HR800 micro-Raman spectrometer (HORIBA, Jobin Yvon, Kyoto, Japan) combined with a laser confocal microscope. To estimate the thickness of the silica film, we used the microscope of the HR800 spectrometer equipped with a ×100 objective. By comparing between the film images obtained with the microscope focused onto the inner and outer film boundaries, we found that each spin coating run formed one to three layers of silica spheres on the wafer. To fabricate SERS substrates,

we used concentrated GNR sols obtained by the redispersion of 12 mg cAMP of GNP powder in 1 mL of distilled water. A drop of a GNR sol of controllable volume was placed on a film of silica spheres on a silicon wafer and dried at room temperature. This process was repeated several times to attain the desired surface and volume densities of the GNRs embedded in and deposited on the OPC film. For comparative purposes, we also fabricated SERS substrates by depositing GNR sols differing in concentration directly on plain silicon wafers as described previously in [33]. Results and discussion Properties of GNR powders Figure 1a shows a TEM image of a GNP nanopowder redispersed in water. The size and shape of the nanoparticles practically do not differ from those the as-prepared GNRs had before freeze-drying. Accordingly, there are no essential differences between the extinction spectra of the samples recorded prior to and after freeze-drying (Figure 1b).