ANALYSIS OF OILS USING FLUORESCENCE.
11
However, no attempt was made to correlate the lifetime data with chemical composition.
In 1987, Abu-Zeid et al. revisited the use of fluorescence lifetimes as a potential tool for
identifying oil slicks.
11 Unfortunately, their lifetime measurements did not include any
pre-exponential factors so it is not possible to assess the true fluorescence lifetimes of the
oils.
The complex composition of crude oils ensures that the measured fluorescence decay
curve is the sum of a distribution of individual decay curves. As such, one needs to
exercise caution in discussing fluorescence lifetimes. Wang and Mullins,
25 in their
detailed study of the fluorescence lifetime of crude oils used a bi-exponential fit to
describe the fluorescence decay curves. More recently, we have used the intensityaveraged lifetime (τ ) calculated using either bi-or tri- exponential fits as a reproducible
and standardised lifetime measurement for crude oils.
10, 21, 22, 23, 32 Figure 8 shows the τ
dependence with emission wavelength for a series of North Sea oils. In general, the
wavelength at which the maximum average lifetime is measured occurs at longer
wavelength for heavy crude oils than it does for light oils.
16, 21, 22 The variation in lifetime
going from 450 nm to ~600 nm emission wavelength is more pronounced for the lighter
oils as is seen from Figure 8. At a particular point, a maximum value for the average
lifetime (τ max ) is recorded at a wavelength λ τmax (e.g. ~ 600 nm for N11 with 380 nm
excitation), which is dependant on the chemical characteristics of each oil. Figure 8 also
shows a noticeable decrease in the average lifetime at emission wavelengths greater than
~600 nm. This ‘curved’ wavelength dependence is due to the complex interplay between
collisional energy transfer and quenching processes.
21, 22
At short emission wavelengths, fluorophores have large bandgaps and can readily
undergo collisional energy transfer with larger, aromatic, small bandgap molecules. The
rate of energy transfer is dependant on the bandgap between the two populations and so
the value of τ relative to the maximum lifetime (τ max ) is shortened the most at short
emission wavelengths where the bandgap is greatest. As the emission wavelength
increases the energy gap between donor and acceptor decreases, and the concentration of
large aromatic acceptor molecules decreases. The first factor results in an increase in the
energy transfer rate but the second factor results in a decrease in the rate, as there are
fewer acceptor molecules. Overall, the effect of both factors results in a net overall
decrease in the rate of energy transfer leading to an increase in τ until τ max is reached at
a wavelength λ τmax . λ τmax is red-shifted as the API gravity of the oils decreases which is
evident from Figure 8.
11
However, no attempt was made to correlate the lifetime data with chemical composition.
In 1987, Abu-Zeid et al. revisited the use of fluorescence lifetimes as a potential tool for
identifying oil slicks.
11 Unfortunately, their lifetime measurements did not include any
pre-exponential factors so it is not possible to assess the true fluorescence lifetimes of the
oils.
The complex composition of crude oils ensures that the measured fluorescence decay
curve is the sum of a distribution of individual decay curves. As such, one needs to
exercise caution in discussing fluorescence lifetimes. Wang and Mullins,
25 in their
detailed study of the fluorescence lifetime of crude oils used a bi-exponential fit to
describe the fluorescence decay curves. More recently, we have used the intensityaveraged lifetime (τ ) calculated using either bi-or tri- exponential fits as a reproducible
and standardised lifetime measurement for crude oils.
10, 21, 22, 23, 32 Figure 8 shows the τ
dependence with emission wavelength for a series of North Sea oils. In general, the
wavelength at which the maximum average lifetime is measured occurs at longer
wavelength for heavy crude oils than it does for light oils.
16, 21, 22 The variation in lifetime
going from 450 nm to ~600 nm emission wavelength is more pronounced for the lighter
oils as is seen from Figure 8. At a particular point, a maximum value for the average
lifetime (τ max ) is recorded at a wavelength λ τmax (e.g. ~ 600 nm for N11 with 380 nm
excitation), which is dependant on the chemical characteristics of each oil. Figure 8 also
shows a noticeable decrease in the average lifetime at emission wavelengths greater than
~600 nm. This ‘curved’ wavelength dependence is due to the complex interplay between
collisional energy transfer and quenching processes.
21, 22
At short emission wavelengths, fluorophores have large bandgaps and can readily
undergo collisional energy transfer with larger, aromatic, small bandgap molecules. The
rate of energy transfer is dependant on the bandgap between the two populations and so
the value of τ relative to the maximum lifetime (τ max ) is shortened the most at short
emission wavelengths where the bandgap is greatest. As the emission wavelength
increases the energy gap between donor and acceptor decreases, and the concentration of
large aromatic acceptor molecules decreases. The first factor results in an increase in the
energy transfer rate but the second factor results in a decrease in the rate, as there are
fewer acceptor molecules. Overall, the effect of both factors results in a net overall
decrease in the rate of energy transfer leading to an increase in τ until τ max is reached at
a wavelength λ τmax . λ τmax is red-shifted as the API gravity of the oils decreases which is
evident from Figure 8.