This paper turned out to be a gold mine of information on fluorescent spectrometry of crude oils. Some of the points that stood out to me were: -visible excitation (>450 nm) is generally not suitable for crude oil excitation -complexity of crude oil results in broad spectrum of adsorption and excitation, meaning identification of specific compounds is not feasible -In general, adsorbed light is red-shifted (i.e. light emitted is of greater wavelength than light adsorbed) -Most emissions for light and heavy crude oils are in the 450-500 nm range -with that said, the wavelength of adsorbed light is important- it looks like the LED black light i'm using generates somewhere between 380 and 400 nm light -Because there are many different compounds in the oil, fluoresced light may be re-adsorbed, and re-fluoresced -It appears that with the black light I am using, approximately 60% of the energy adsorbed by any oil will be emitted in the form of fluorescence
I forgot to mention this on my last update, but the spectra I posted of the detergent fluorescing, I had to use a 4 second exposure time when taking the pictures to get usable spectra.
Another issue that may come up is that when looking at fluorescence of real water samples, particulate matter may scatter the black light, creating a spectrum including the black light wave lengths, which would make the sample look like it was fluorescing, when it may not have been. One way to solve this would be use a purely UV (non visible) light source. Another could be to throughly filter the samples. I'm not sure if the spectra of the UV light source could be distinguished reliably from the fluorescent spectrum.
Also this week, I will hopefully calibrate the spectrometer by using a salt flame test, which should produce a well known spectra to compare to.