Fluorescence spectroscopy is a highly sensitive and specific
method for oil identification since petroleum oils contain a mixture
of many fluorescent aromatic hydrocarbons. Because of this complex
mixture, the spectral distribution and the intensity of the
fluorescence will vary with the chemical composition of the oil.
Therefore at fixed instrumental conditions and a given concentration
of oil, the fluorescence spectrum results in a characteristic
"fingerprint" of a given oil.
The first part of this work involved the development of a library
of fluorescence spectra for model aromatic compounds common in
environmental samples. The techniques utilized were based on
synchronous fluorescence whose advantages in oil identification have
been demonstrated. Extensive studies were carried out using the
relatively new and little used technique of variable angle synchronous
fluorescence (v.a.s.f.). Non-linear v.a.s.f., where the angle of scan
trajectory is altered during a scan. was developed as a means of
obtaining desired "sections" through the excitation-emission matrix.
Good selectivity and calibration linearity were observed for mixtures
of compounds. The results obtained demonstrate the potential of the
technique in oil identification. Derivative spectroscopy was applied
to enhance minor spectral details in the comparison of closely related
The second part of the work involved the application of the
previous techniques to real-world samples in the form of crude oils
and tar balls. The "fingerprinting principle" is based on the fact
that environmental samples of oils can be correlated with possible
sources by comparing their stable compositional features. Hence
fluorescence studies were performed on several oil samples before and
after laboratory weathering. By observing spectral regions where
weathering occurs, it was noted that these fluorescence techniques for
oil identification could be further improved.
Effects of cyclodextrins and micelles. and the process of
deoxygenation by nitrogen bubbling and sodium sulphite addition, were
investigated as further techniques for discriminating between similar
samples and enhancing certain spectral details.
Finally, other techniques that have been used in oil
identification such as infra-red, thin-layer chromatography, and gas
chromatography were briefly examined and their use in a multi-method
oil analysis approach discussed.
A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of Doctor of Philosophy of Loughborough University.