Purity analysis with HPLC-PDA

[Perreman]

Hi everyone.
A customer sent us a sample which we were supposed to analyze regarding its purity using a PDA detector. We do not know which substance this is, only that there is supposed to be a single substancein it. I was told to perform this the same way as a prior sample from the same customer. THe problem I have is the evaluation of the results.
The sample was blanksubtracted. The overview of the results show that there were some impurities at several wavelengths. These wavelengths with impurities was used to report the main peaks area% relative to the total peak area, simply showing the customer that their substance in the sample had the purity displayed as area% of the total area at these different wavelengths.

Is it possible to get a total purity somehow from doing an analysis with a PDA detector? Or is there any other way to get a total purity of a sample when not knowing anything about the characteristics of the sample substance?

Any hints or suggestions are very much welcome!

[tom jupille]

Is it possible to get a total purity somehow from doing an analysis with a PDA detector? Or is there any other way to get a total purity of a sample when not knowing anything about the characteristics of the sample substance?

The short answer is "No".

The longer answer: the PDA can use two sources of difference: chromatographic and spectrophotometric. If the impurities can be completely resolved from the product by LC, then compounds with identical spectra can be quantitated. If the impurities have very different absorbance spectra, then compounds with identical chromatographic retention can be quantitated. And, of course, any situation in between (some separation and some spectral difference).

In practice, that means you have to know *something* about the sample: what are the potential impurities (synthesis intermediates? degradation products?) and how are they likely to affect the chromatographic and spectrophotometric properties (same chromophore, but isolated from the reaction site of the degradation? changes to the chromophore?).

The problem is that UV spectra in solution are notoriously "information-poor"; most of them are blobs with very little structure. If you want to quantitate "in the dark", LC-MS is you best bet,