Hadamard-Transform Fluorescence Excitation-Emission-Matrix Spectroscopy: Rapid 3-Dimensional Fluorescence Data for Analytical and Industrial Applications

Abstract

Fluorescence spectroscopy is a very sensitive and selective analytical technique that can be used to monitor and quantify a sample. The method is non-destructive and creates no chemical waste. In conventional fluorescence spectroscopy, a sample is excited with a single wavelength of light and the emitted spectrum is collected. The conventional method is well suited for simple samples and sample mixtures, but it fails when several fluorophores emit similar spectra or have a low absorption cross section at the excitation wavelength. Excitation-emission-matrix (EEM) fluorescence spectroscopy addresses both shortcomings by using many excitation wavelengths and recording a complete emission spectrum at each excitation wavelength. This creates a two-dimensional spectrum, which may be displayed as a contour plot or “heat map”. Fluorophores produce “hills” in these EEM spectra and may be identified and quantified using multivariate data analysis methods.

Conventional EEM spectrometers scan the excitation light sequentially through a region of the UV-Vis spectrum while recording the emission spectra of the fluorophore with a spectrometer; this process may take up to an hour for a well-resolved spectrum. Not surprisingly, EEM spectroscopy has previously not been used for kinetic studies or the detection of transient species and intermediates. In this thesis, a new Hadamard-transform excitation-emission-matrix (HT-EEM) spectrometer is described and developed, which reduces the data acquisition time from 30-60 minutes to under 140 milliseconds. A new data acquisition algorithm (Super-Cycle algorithm) was applied to a HT-EEM dataset, and it demonstrated an even higher time resolution at less than 20 ms/spectrum. As a result, analytical problems that were intractable only a few years ago are now within reach. For example, with our rapid analysis time we can perform on-line and real-time analysis of samples, such as turbine oils, and check their molecular composition and quality. Our new HT-EEM technique was also applied to existing analytical methods such as HPLC separation, where we can record the EEM spectrum of the different fluorophores within a sample.