Nanoscale materials have great applications in many areas. One of these applications is for manufacturing ultra-compact and efficient sensors for chemical and biological molecule detection. Noble metals, such as gold (Au) and silver (Ag), because of their distinguished optical property"localized surface plasmon resonances (LSPRs) that exhibit low loss, are ideal materials to fabricate these nanoscale plasmonic particles or structures. This work addresses the synthesis, characterization, and sensing applications of Au and Ag nanoparticles (NPs). The progress on certain subjects related to our work"NP synthesis, surface functionalization, Au sphere-film structure and two-photon fluorescence"are reviewed in Chapter 1. We also show the calculation results of LSPRs of Au nanosphere suspensions using Mie theory. The measured extinction spectra of Au nanosphere suspensions agree with the calculated results very well. Chapter 2 is a chapter describing the chemical synthesis of a variety of NPs, such as Ag prisms and cubes, Au spheres, rods, and bipyramids. These experiments involved different synthetic mechanisms and methods which enabled us to prepare NPs with desired shapes and optical properties. To put these NPs into application, it is desirable and sometimes necessary to functionalize their surfaces. In Chapter 3, we present the functionalization of Ag cubes with poly(allylamine hydrochloride) (PAH) and poly(allylamine hydrochloride)-dithiocarbamate (PAH-DTC), which follows our previous work on Au NPs. The purpose of studying Ag instead of Au is to use the stronger plasmonic enhancement in Ag when applied to two-photon imaging applications. However, we found that PAH-DTC shrank the Ag cubes. We also functionalized the cationic hexadecyltrimethylammonium bromide (CTAB)-stabilized Au NRs with anionic poly(sodium 4-styrenesulfonate) (PSS). Coated with the strong polyelectrolyte PSS, the NRs become more manageable and can be stable for over six months and are easily immobilized onto positively charged substrate. We put PSS-functionalized Au NPs into use and studied their adsorption process onto PAH-coated optical fiber tapers by monitoring the transmission light through the fiber. When the diameter of the fiber taper gets smaller, stronger coupling occurred between transmitted light inside the taper and the Au NPs on the taper surface (cylinder). This coupling resulted in a loss of the guided light at the plasmon resonance wavelength of the NPs. By monitoring this loss, we can study the adsorption rate of Au NPs onto the fiber. In Chapter 4, we used Au nanospheres to study the adsorption rate on substrates with different curvatures. We also established a theoretical model to explain this phenomenon for cylindrical surface as well as planar and spherical surfaces. Our results fit well with the theory, which predicts that particle adsorption rates depend strongly on surface geometry, and can exceed the planar surface deposition rate by over two orders of magnitude when the diffusion length of the particle is large compared to the surface curvature. In Chapter 5, we studied the optical properties of Au nanospheres separated from a thick Au film by a polyelectrolyte multilayer (PEM) film assembled from PAH and PSS under specific pH condition. The PEM film undergoes swelling and shrinking when the environmental pH is changed as a result of charging and discharging of the polyelectrolytes. Therefore, the PEM film provides an efficient means to tune the distance between Au spheres and Au film. The extinction peak blue-shifted as much as 100 nm when the pH of the water changed from pH 10 to pH 3 for 100 nm diameter Au spheres on a PEM film assembled at pH 9.5. Our preliminary estimates that the gap between sphere and surface can be as small as a few nm even though the film itself is tens of nm thick when it is not constrained by Au spheres. We studied two-photon excitation fluorescence (TPEF) from Ag triangles in Chapter 6. The triangles were fabricated by nanosphere lithography, which used convective self-assembly to make the nanosphere mask. The LSPRs of the nanotriangles were tuned to be in the 800--900 nm range to match with the Ti:Sapphire pulse laser at 880 nm. We found that certain spots on the fluorescence images gave rise to larger fluorescence intensity than rest of the area. SEM imaging reveals that the unusually bright spots seen on the surface were related to regions where the triangles transformed to spherical particles. The larger intensity is tentatively ascribed to the plasmon resonance of those spherical particles in ~400 nm range. / Ph. D.

Optical sensor systems that integrate Long-Period-Gratings (LPG) as the detection arm have been proven to be highly sensitive and reliable in many applications. With increasing public recognition of threats from bacteria-induced diseases and their potential outbreak among densely populated communities, an intrinsic, low-cost biosensor device that can perform quick and precise identification of the infection type is in high demand to respond to such challenging situations and control the damage those diseases could possibly cause. This dissertation describes the development of a biosensor platform that utilizes polymer thin films, known as ionic self-assembled multilayer (ISAM) films, to be the sensitivity- enhancing medium between an LPG fiber and specific, recognition layer. With the aid of cross- linking reactions, monoclonal antibodies (IgG) or DNA probes are immobilized onto the surface of the ISAM-coated fiber, which form the core component of the biosensor. By immersing such biosensor fiber into a sample suspension, the immobilized antibody molecules will bind the specific antigen and capture the target cells or cell fragments onto the surface of the fiber sensor, resulting in increasing the average thickness of the fiber cladding and changing the refractive index of the cladding. This change occurring at the surface of the fiber results in a decrease of optical power emerging from the LPG section of the fiber. By comparing the transmitted optical power before and after applying the sample suspension, we are able to determine whether or not certain bacterial species have attached to the surface of the fiber, and as a consequence, we are able to determine whether or not the solution contains the targeted bacteria. This platform has the potential for detection of a wide range of bacteria types. In our study, we have primarily investigated the sensitivity and specificity of the biosensor to methicillin- resistant Staphlococcus aureus (MRSA). The data we obtained have shown a sensitive threshold at as low as 102 cfu/ml with pure culture samples. A typical MRSA antibody-based biosensor assay with MRSA sample at this concentration has shown optical power reduction of 21.78%. In a detailed study involving twenty-six bacterial strains possessing the PBP2a protein that enables antibiotic resistance and sixteen strains that do not, the biosensor system was able to correctly identify every sample in pure culture samples at concentration of 104 cfu/ml. Further studies have also been conducted on infected mouse tissues and clinical swab samples from human ears, noses, and skin, and in each case, the system was in full agreement with the results of standard culture tests. However, the system is not yet able to correctly distinguish MRSA and non-MRSA infections in clinical swab samples taken from infected patient wounds. It is proposed that nonspecific binding due to insufficient blocking methods is the key issue. Other bacterial strains, such as Brucella and Francisella tularensis have also been studied using a similar biosensor platform with DNA probes and antibodies, respectively, and the outcomes are also promising. The Brucella DNA biosensor is able to reflect the existence of 3 Brucella strains at 100 cfu/ml with an average of 12.2% signal reduction, while negative control samples at 106cfu/ml generate an average signal reduction of -2.1%. Similarly, the F. tularensis antibodies biosensor has shown a 25.6% signal reduction to LVS strain samples at 100 cfu/ml, while for negative control samples at the same concentration, it only produces a signal reduction of 0.05%. In general, this biosensor platform has demonstrated the potential of detecting a wide range of bacteria in a rapid and relatively inexpensive manner. / Ph. D.