Design, operation and applications of a visible-light confocal scanning Fourier transform Raman microscope for volumetric Raman spectrochemical imaging

A new type of confocal Raman microscope called a Fourier transform confocal Raman microscope (FT-CRM) was designed, built and characterized with respect to its spatio-spectral imaging properties. Several different applications of the FT-CRM are presented that take advantage of its unique spectral and spatial imaging characteristics. The instrument combines focused illumination with spatially-filtered detection in a confocal optical configuration to collect photons scattered from a diffraction-limited volume in the sample (typically ${<}5 times10 sp{-18} m sp3)$ and reject photons from outside that region. The molecular vibrational information encoded in the inelastic, or Raman, spectral component of light scattered from the confocal volume is measured with a visible light Fourier transform Raman spectrometer. By scanning the sample relative to the confocal volume, a volumetric Raman spectrochemical image of the sample can be constructed.Raman scattering is an inherently inefficient process; hence an optimal radius pinhole must be found that balances the FT-CRM optical throughput against the microscope spatial resolution and image contrast. Detailed experimental measurements mapped out the FT-CRM spatial response (axial and lateral), optical throughput and image signal-to-background and signal-to-noise ratios as a function of pinhole radius. Excellent agreement was found between these measurements and the predictions of a theoretical microscope model also developed as part of this thesis. Several applications of the FT-CRM included volumetric compositional imaging of three-dimensional chemically inhomogeneous materials such as cellulose and polyester fibers in water or two immiscible optically-similar liquids, water and trichloroehthylene, in a porous quartz sandstone matrix. The potential of the FT-CRM for non-invasive spectrochemical detection and imaging through a turbid tissue-like medium was demonstrated and a new spectral estimator, Fast Orthogonal Search, was evaluated to replace the discrete Fourier transform to improve the microscope performance