Passive mass transport for direct and quantitative SERS detection using purified silica encapsulated metal nanoparticles

This thesis focuses on understanding implications of nanomaterial quality control and mass transport through internally etched silica coated nanoparticles for direct and quantitative molecular detection using surface enhanced Raman scattering (SERS). Prior to use, bare nanoparticles (partially or uncoated with silica) are removal using column chromatography to improve the quality of these nanomaterials and their SERS reproducibility. Separation of silica coated nanoparticles with two different diameters is achieved using Surfactant-free size exclusion chromatography with modest fractionation. Next, selective molecular transport is modeled and monitored using SERS and evaluated as a function of solution ionic strength, pH, and polarity. Molecular detection is achieved when the analytes first partition through the silica membrane then interact with the metal surface at short distances (i.e., less than 2 nm). The SERS intensities of unique molecular vibrational modes for a given molecule increases as the number of molecules that bind to the metal surface increases and are enhanced via both chemical and electromagnetic enhancement mechanisms as long as the vibrational mode has a component of polarizability tensor along the surface normal. SERS signals increase linearly with molecular concentration until the three-dimensional SERS-active volume is saturated with molecules. Implications of molecular orientation as well as surface selection rules on SERS intensities of molecular vibrational modes are studied to improve quantitative and reproducible SERS detection using internally etched Ag@Au@SiO₂ nanoparticles. Using the unique vibrational modes, SERS intensities for p-aminothiophenol as a function of metal core compositions and plasmonics are studied. By understanding molecular transport mechanisms through internally etched silica matrices coated on metal nanoparticles, important experimental and materials design parameters are learned, which can be subsequently applied to the direct and quantifiable detection of small molecules in real samples without the need for lengthy separations and assays.