Chromatography Forum

Hi

I am having trouble with a negative dip in my HPLC chroamtography. It occurs arounf 29minutes at all the injections, and I have changed absolutely nothing with my method? I am currently flushing 20% Ethanol and Water through the column, can you suggest anything else that might remove this contamination?

Thanks,
Barbara

Does this dip interfere with your peak(s) of interest?

Does this peak appear if you do a blank run?

Does this peak appear if you do an injection of water or methanol?

Thanks for your reply!
The negative peak occurs after my peak of interest. It is not the column, as I have taken that off and performed a direct injection. It does appear in a blank run.
I have tried cleaning the system with IPA/Water mix, and still no joy?!

What kind of samples are you running, when was the last time you PM'd the pump/ALS, and maybe you should try washing with stronger solvents (just gradually increase their gradient so they remain miscible).

The key is in CPG's last question (injection of water or methanol) with one addition: what happens if you inject mobile phase.

What all of this is leading up to is that what you are describing sounds very much like a "syatem peak". More details about your mobile phase and detection might help to pin it down further. Another (more remote) possibility is an artifact from the "refernce wavelength" function of a DAD, but the fact that it shows up in a blank makes it less likely.

In any case, if it doesn't interfere with a peak of interest, you can always deal with it the old-fashioned way: ignore it.

The key is in CPG's last question (injection of water or methanol) with one addition: what happens if you inject mobile phase.

What all of this is leading up to is that what you are describing sounds very much like a "syatem peak". More details about your mobile phase and detection might help to pin it down further. Another (more remote) possibility is an artifact from the "refernce wavelength" function of a DAD, but the fact that it shows up in a blank makes it less likely.

In any case, if it doesn't interfere with a peak of interest, you can always deal with it the old-fashioned way: ignore it.

I had the same problem as the OP. The negative dip can be seen towards the end of the chromatogram, and I have not changed my methods. This phenomenon is not column related (seen it on a new column).

My answer to CPG's questions are: sometimes Yes (the HPLC is used for several methods), Yes, and Yes (methanol) . Haven't tried Tom's additional question though.

Flushing the system with MeOH seems to have no effect, and this effect is more prominent at lower wavelength (225nm) than higher wavelengths (I am using a PDA/DAD detector). The reference wavelength or Blank subtraction is not enabled.

My mobile phase is comprised of 0.1% TFA in ACN and 0.1% TFA in H2O. The instrument is a Waters Alliance 2695 with 2998 PDA detector.

It's probably just a normal gradient artifact. These usually go upwards in RP systems, but if your ACN gets old or something is not as thoroughly degassed as you are used to, they can go negative. It may just be a matter of cracking open fresh ACN and TFA bottles. If it is not interfering with your peaks of interest, just ignore it.

Hey DR, exactly WHY does old acetonitrile/TFA cause the baseline to drop negatively in gradients? I have done lots of reverse phase testing and have seen the baseline rise, and sometimes fall (all with good chromatography) but never knew why it did that.

No I never got the issue of negative peaks sorted I just took Tom's advise and ignored them, as they did not interfere with my peaks of interest! Sorry I cant be of more help.

Barbara

TFA apparently forms a charge transfer complex with ACN such that the actual absorbance spectrum shifts. If I recall, ther eis an isosbestic point on the spectrum at 214 nm. Se this article by Rozing et al:
http://tinyurl.com/7gg8hht

we isolated it to one of the solvents in our Mobile phase. We changed vendors and are not seeing the dip anymore. YEAH!!

Barbara et. al. - if you see negative dips, investigate whether it is occuring on more than one method. That will help distinguish whether it is "method" related or possibly system related. If the negative dip is smooth akin to s solvent front dip, then it probably is a system peak of some kind. We had instance once where there was a straight up and down dip (very sudden not smooth like a solvent front dip) that only occured with gradient analyses. After many months of me moaning to the instrument vendor that it was related to the detector they found a defective main board. Why it only occured with gradient methods I have no idea. I doubt that issue would have been picked up following a troubleshooting guide