If you haven't run blank gradients, then you don't know whether what you think are peptides actually are or whether they are "ghost peaks" due to impurities. As several here have advised, you should first run a blank gradient - a run without an injection. Any peaks are most probably due to impurities in the water, although I've occasionally had trouble with TFA impurities as well. Also, it sounds as if you're using a gradient with TFA in the aqueous phase only, so even a change in the molar absorptivity of the TFA is possible. Also, as David Soriano mentions, I've frequently encountered absorbance changes during the (usually fast) reverse gradient back to initial conditions.

Run a gradient without an injection, then run with an injection of the sample solvent, THEN - if everything is working - run your sample.

What about pressure? I see usually changes in baseline at 268 nm because of pressure changes.

Did you check the column pressure? It would be a pressure drop in the system.

good luck.

francesco

Run a blank! If you see the same effect it will be the gradient effect using this mobile phase system. If you use same concentration of TFA in both ACN and Water phase this might stop the drift.

A blank run will help you.

I've observed the same effect using an almost identical mobile phase (gradient ACN/water + TFA). Only the water is acidified with TFA (0.1 %). TFA absorbes at low wavelength (210 nm for me).

Good luck!

Does it coincide with the column conditions going back to initial settings for column reconditioning?

If you haven't run blank gradients, then you don't know whether what you think are peptides actually are or whether they are "ghost peaks" due to impurities. As several here have advised, you should first run a blank gradient - a run without an injection. Any peaks are most probably due to impurities in the water, although I've occasionally had trouble with TFA impurities as well. Also, it sounds as if you're using a gradient with TFA in the aqueous phase only, so even a change in the molar absorptivity of the TFA is possible. Also, as David Soriano mentions, I've frequently encountered absorbance changes during the (usually fast) reverse gradient back to initial conditions.

Run a gradient without an injection, then run with an injection of the sample solvent, THEN - if everything is working - run your sample.

Dear colleague,

the gradient chromatography at so large range of organic solvent concentration (10-80%ACN) at low wave length is complicated: some impurities in the mobile phase retain in the column at low ACN concentration, and during the gradient they are eluted (low baseline when column is washed), and after conditioning at low ACN concentration they (impurities) leave the column with delay (increasing the baseline), till equilibration. Particular elimination of this effect could be achieve using very pure mobile phase components, dissolving the sample in the start mobile phase and use gradient in reduced range of ACN concentration. The best way - to use ELSD (evaparative light scattering detector).

Regards and ciccess.

Dr V. Bardarov

Dear All

Many thanks for all insights and suggestions. I have injected blank twice, and indeed the pattern (of negative 215-220nm abosoption) is repeated, however not my peptide peaks. I am quite sure I see the peptides in the right place. Apparently, when ACN % is beyond 37% column pressure starts increasing and this causes steep absortion decrease at 215/220. I will try remaking all solutions and also test in a Shimadzu system for confirmation. As I am inexperienced, I am quite lost at this observation. Will try to post the result here later.

Dear Bardarov,

You may indeed have a point there, I will think it over. However the disturbances do not seem to be leaving impurities as the variations do not peak. Also I am using MilliQ water and HPLC-grade solvents. I had never heard of ELSD, this is quite interesting! Yet probably out of our monetary reach, though...

Many thanks!

Dear Dethier: "A blank run will help you.

I've observed the same effect using an almost identical mobile phase (gradient ACN/water + TFA). Only the water is acidified with TFA (0.1 %). TFA absorbes at low wavelength (210 nm for me)." -- YES! This is exactly what I am talking about! How did you circumvent the problem? I will try to post the result, and you can see if it looks familiar.

Dear Soriano: "Does it coincide with the column conditions going back to initial settings for column reconditioning?" -- Indeed from monitoring the column while it returns to initial state the event slowly returns to initial conditions. I am pretty sure the phenomenon is due to solvent change in the column, resulting in increased pressure and different absorbance. However I am not sure why previous authors with same protocol did not experience the same, according with their papers.

I think it is artifact. generally at 215-220nm absorption chromotogram gives a negative miminima at 5-10min range initially (a sharp minima or some time broad) then chromotogram goes fine without any disturbance...

Dear Eduardo,

I didn't circumvent the problem as my peaks are eluted before the baseline changes. The effect of ACN (without TFA) impacts the baseline AFTER the separation in my case.

If you can't ignore the problem, you can try to substract your blank chromatogram from your sample chromatogram (the software can do it), but I would avoid this technique as much as possible.

I wish you much succes!

Dear Berenice, also in my case peaks eluted before the disturbance. However I cannot be sure that nothing came out during it. I have been indicated a solution, will try it out today and post here, best wishes!

If your two mobile phases include 0.1% TFA, the Abs shift is likely due to your ACN. Even HPLC-grade ACN can absorb at the low end. Few thoughts:

Try a different lot and/or manufacturer of your HPLC-grade (or purer) ACN, as lots and manufacturers can vary. Look at the specifications between manufacturers to see if they differ in Abs at the low end.

Degas both mobile phases (sonicate the source bottles under vacuum) every morning before use. This may reduce your pressure and lower baseline interference.

Try fresh TFA, as TFA oxidizes (or otherwise degrades) upon opening. We buy small 50 mL TFA bottles, and discard them 1 month after opening (at RT). If nothing else, fresh TFA will likely give you sharper elution peaks.

Thanks Carl, this is really good advice! I am trying a few modifications right now...

Dear all, I have found the answer! Many thanks to everyone and for all extra advices. I am taking notes of all. In the end, I am even embarassed to say, but it was my inexperience with the software. Upon setting the ACN % gradient protocol, I assumed the software would set my column to the initial condition of low % ACN. I was wrong, and it was injecting 10% ACN into 80% ACN directly when the run began, and slowing increasing. This abrupt change in solvent concentration was causing the observed conditions, also with abrupt pressure alteration. I was trying new solutions and even changing run times, and I finally realized the problem while monitoring the run closely. After setting the initial 0 condition to 10% ACN, the problem is resolved. Thanks again for all help!

Also, this was really helpful: http://www.waters.com/webassets/cms/library/docs/wa20769.pdf