Monodisperse, citrate-stabilized gold nanoparticles of sizes ranging from 15 to 40 nm were synthesized and characterized by small angle X-ray scattering and UV-vis experiments. Identical surface properties of nanoparticles of different sizes to avoid variation in the chemical surface-enhanced Raman scattering (SERS) enhancement, as well as selection of experimental conditions so that no aggregation took place, enabled the investigation of enhancement of individual nanospheres. Enhancement factors (EFs) for SERS were determined using the dye crystal violet (CV). EFs for individual gold nanospheres ranged from 10 2 to 10 3 , in agreement with theoretical predictions. An increase of the EFs of individual spheres with size can be correlated to changes in the extinction spectra of nanoparticle solutions. This confirms that the increase in enhancement with increasing size results from an increase in electromagnetic enhancement. Beyond this dependence of EFs of isolated gold spheres on their size, EFs were shown to vary with analyte concentration as a result of analyte-induced aggregation. This has implications for the application of nanoparticle solutions as SERS substrates in quantitative analytical tasks.

Surface-enhanced Raman scattering (SERS) labels and probes consisting of gold and silver nanoaggregates and attached reporter molecules can be identified by the Raman signature of the reporter molecule. At the same time, SERS hybrid probes deliver sensitive molecular structural information on their nanoenvironment. Here we demonstrate full exploitation of the multifunctional and multiplexing capabilities inherent to such nanoprobes by applying cluster methods and principal components approaches for discrimination beyond the visual inspection of individual spectra that has been practiced so far. The reported results indicate that fast, multivariate evaluation of whole sets of multiple probes is feasible. Spectra of five different reporters were shown to be separable by hierarchical clustering and by principal components analysis (PCA). In a duplex imaging approach in live cells, hierarchical cluster analysis, K-means clustering, and PCA were used for imaging the positions of different types of SERS probes along with the spectral information from cellular constituents. Parallel to cellular imaging experiments, cytotoxicity of the SERS hybrid probes containing aromatic thiols as reporters is assessed. The reported results suggest multiplexing applications of the nontoxic SERS nanoprobes in high density sensing and imaging in complex biological structures.