Novel Imaging Systems: Multivariate Optical Computing in the UV-VIS

Abstract

Imaging Multivariate Optical Computing (MOC) is a powerful analytical tool that combines digital imaging and predictive spectroscopy without the use of a spectrometer. Traditional techniques for multivariate analysis are both time consuming and costly involving expensive multichannel instruments, difficult computations on large data sets and confinement to the laboratory. By encoding an angle-tolerant spectral pattern specific to a target analyte onto an interference filter, a snapshot of the chemical distribution and concentration can be obtained quickly and inexpensively.This paper summarizes the advances in chemical imaging utilizing angle-tolerant Imaging Multivariate Optical Elements (IMOE) from the UV to visible regions of the spectrum using uncollimated sources. Two applications are demonstrated using both transmission and diffuse reflectance data. These optical elements are fabricated via reactive magnetron sputtering of alternating layers of high and low refractive index materials, (Nb2O5 and SiO2, respectively). A single CCD camera in a swivel design rotating about the IMOE filter axis is employed for both transmittance and reflectance imaging. Our current UV diffuse reflectance technique miniaturizes the chemical imaging prototype with a catadioptric image-splitting prism block and a single monochrome camera controlled by a handheld computer.