Chromatography Forum

In an HPLC method I am currently developing I am observing an odd phenomenon:

The peak of the active as such is perfectly symmetrical (USP Tailing 1.04).
However, when scanning over the peak, recorded with a PDA detector in the range 190-400 nm, the spectrum changes quite significantly in the wavelength region up to approx. 250 nm. At the beginning and end of the peak, the spectrum would have a maximum absorbance at approx 200nm, whereas in the centre of the peak, the maximum of the band is 242nm.
I am quantifying the peak at 301nm, a maximum of another absorption band which is unchanged during the elution of the peak.
The molecule contains an indole system, a sulfonamide structure, and a carbamate structure. The mobile phase is acetonitrile/ammonium formiate buffer pH 2.8. The standard solution contains additionally, to support dissolution, 10% DMF.

Could anyone explain the symmetrical changes in the spectra throughout the peak elution, also there is no indication that it would be a mixture of different compounds?

Thank you very much!

There are other molecules--perhaps unrelated to your target analyte in question--that absorb in the 220 to 240 range. You need to compare those peaks with your reference standard. Where do the peaks come out on your reference standard? Also, what wavelenght does the manufacturer say it absorbs at?

I would guess that your spectral mobile phase background reference points aren't being subtracted correctly, and what you are seeing at the beginning and end represents mobile phase absorption.

Please keep aving fun,

Bruce Hamilton

Thanks for the answers.

By re-inspection of the chromatograms/spectra I would agree that it is indeed an issue with the mobile phase background. So I can see the absorbance at approx. 200nm also in the baseline, up to 50mAbs.
As a I quantify the peaks at 301nm that should not be a real problem.

I will have a look into it in the next run, maybe using another HPLC.

The fun continous!

Whereas running spectra like you did, without subtracting the background (mobile phase, etc.) is useless, it doesn´t effect your quantitation at the analytes max. wavelength unless you are working outside the useful absorbance range of your detector.

If yours is an isocratic method, I would suggest choosing your reference points so that they're a little farther from your peak of interest. If it is a gradient method, this won't help as much.

If relocating your reference points doesn't help it may be time to entertain the possibility of a coeluting analyte. Jumpshooter's suggestion would go a long way toward confirming this (as would a MSD, but I don't think your MP is MSD compatible).