Solvent perturbation spectroscopy and multiplexed concentration quantitation of SERRS

This work has been aimed at solving chemistry related problems with multivariate data analysis methods and Raman spectroscopy. The first chapter describes spectroscopic investigations of the solvent perturbations around solute molecules. Multivariate curve resolution (MCR) has been used to deconvolve a set of Raman spectra with varying solute concentrations to bulk solvent spectrum and the solute-correlated spectrum. As an illustration of the method, we have extracted the solute-correlated spectra of Acetonitrile and cyclo-Hexane in 1,2-Dichloroethane which reveal changes of gauche/trans equilibrium in the solvation-shell around these polar and non-polar solutes. Perturbations of water around small organic solutes and ionic salts have also been studied. The results reveal a larger contribution to the perturbation comes from polar groups of the organic solutes. Results for aqueous nitrile and alcohol solutions with different hydrocarbon chain lengths suggest that water molecules around non polar groups are relatively unperturbed. This observation is supported by the calculated number of perturbed water molecules per solute which are much smaller than the total number of water molecules in the solvent-shell, as revealed by Molecular Dynamics and Neutron diffraction. Water perturbation around alkali halide salts are dominated by water around the anions, as also indicated by previous studies. Our results confirm the weaker hydrogen bonding as the size of the anion increases, virtually no perturbation from water around cations, and the full width of half maximum (FWHM) of perturbed water OH band decreases as the anion size increases. The observed number of perturbed water molecules per anion are in general agreement with previously reported estimates. Polarized Raman studies have also been conducted. The second chapter describes studies which demonstrate a surface enhanced resonance Raman scatteritn (SERRS) based method for reliably quantifing the concentrations with isotopically substituted Rhodamine 6G (R6G) dyes. We have successfully used Partial Least Squares (PLS) analysis to simultaneously quantify two and three dyes which have very similar spectral signatures, in the nanomolar range. Studies of independent validations sets indicate the independence of the concentration prediction on colloid preparation procedure and total concentration