Can I expect fluorescence?

[Mattias]

http://en.wikipedia.org/w/index.php?tit ... svg&page=1

This molecule and the correspodning impurities where either one or two of the sulfate groups are removed. I do not have a scanning FL instrument anymore, so I cannot just measure it.

[tom jupille]

I'm not a spectroscopist by any means, but off the top of my head I would say "no". In order for a molecule to fluoresce, it usually needs a fairly rigid structure so that there are not a lot of vibrational energy levels between the electronic states.

[Mattias]

Thanks Tom,

I think you are right, and will pursue with UV instead. This molecule has a very simple degradation pattern (only two impurities formed), and a FL method would have made life easier (in a complex sample matrix)

[danko]

I’m not for a second in doubt that this molecule it will fluoresce.
A scanning spectrometer would have been the perfect tool to use here, but since you don’t have that, you could proceed as follows:
If you know the absorbance λ max (not under 220 nm) this should be your excitation wavelength. Here I would expect it to be between 230 and 280 nm.
The emission wavelength is most often between 60 and 120 nm above the excitation wavelength. So, here you’ll have to test 3 – 4 different values with 30 – 40 nm long steps, starting from the lowest offset from your excitation/absorption λ max .

Best Regards

[Mattias]

Hi Danko,

I have run DAD on all three substances (mother substance and two impurities), and they have strikingly similar UV-spectra. There are three UV max (206, 225 and 263 nm) in the current mobile phase.

In the product, the drug is mixed with natural flavours that give a terrible baseline. The pattern also is different for every new batch, making it very hard to understand what are impurites and what are peaks from placebo. This is the reason for searching for a more specific detection technique.

Acc. to your suggestion, I should try to set the excitation wavelength to 226 nm and the emission wavelength to something in the range of 280-350 nm. Or should I go for the 263 nm absorption for excitation instead?

[danko]

Acc. to your suggestion, I should try to set the excitation wavelength to 226 nm and the emission wavelength to something in the range of 280-350 nm. Or should I go for the 263 nm absorption for excitation instead?

The lower excitation λ you choose the greater risk for background quenching (i.e. the other things in the sample and even more importantly the mobile phase will absorb a greater deal of the incident energy/light as well as the fluorescence radiation /light). So, if both the initial product/analyte and its degradants exhibit reasonable absorbance at 263 nm it could be the best choice – all depending on the mobile phase. On the other hand – and especially if sensitivity is important, one should keep in mind that the lower wavelength you choose the higher energy is applied on the molecule thus more sensitivity can be achieved.
Might be a good idea to try both wavelengths as an optimisation step.

Best Regards