[clarkstill]

Hi guys, I wonder if any of you have much experience of running fluorescence spectra? I have none, so I'm hoping someone can help me out. Basically I'm trying to analyse the conformation of a molecule containing pyrenes by fluorescence quenching though excimer formation. I'm running the spectra in (spectrophotometric grade) chloroform, but I've run into some contamination issues: when I run the blank spectrum (i.e. no sample, just CHCl3) I'm getting a cluster of peaks with maxima at 377 and 397 nm. Does anyone recognise this contaminant? I can't work out where it's coming from - the solvent has only been stored in glass containers (no polymers etc) and I pipette it directly into the cuvette. I ran H2O too, to make sure it wasn't the cuvette itself that was contaminated, and this gave a clean blank spectrum. Any help appreciated!

[Arkcon]

When doing fluorescence spectra, the excitation wavelength is just as important as the emission.  So you'll have to tell us that.    You'll have to make sure that chloroform doesn't have an excitation at your chosen wavelength, some references will help you here. 

Its good that you've checked the cuvettes and instrument for scattering, so that's out of the way.

You also have to consider Raleigh scattering:  here's a crappy explanation -- https://en.wikipedia.org/wiki/Rayleigh_scattering#From_molecules 

And Raman scattering:  https://en.wikipedia.org/wiki/Raman_scattering  Now this one is a little easier to detect:  the longer the excitation wavelength, the shorter the emission wavelength.  If you raise the excitation slightly, and get the same structure of peaks just at a shorter wavelength, then that's what's happening.  You'll simply have to try to avoid wavelengths where  Raleigh and Raman scatering occur,

[clarkstill]

Thanks for this, the excitation wavelength was 344 nm and the unidentified peaks were invariant when this wavelength is changed (except intensity changes). I think I'm just going to dodge the issue and use acetonitrile

[Corribus]

In the future, please post spectra - the shape of the features can tell a lot about the origin of rogue peaks. Also, acquiring excitation spectra is always a good step in diagnosing a problem.