Change in retention time and baseline
Chromatography Forum: LC Message Board: Change in retention time and baseline
By dan on Wednesday, July 7, 2004 - 02:33 pm:

I've been monitoring several reactions over the past few weeks with HPLC. Suddenly and with the same method, my spectra look very different. Retention times are much longer and there is much more noise in the baseline. The same sample run on another machine, but with the same method produces the "correct" spectrum. Can this be an indication that the column needs cleaning or replacing?

By tom jupille on Wednesday, July 7, 2004 - 04:59 pm:

Just as a point of terminology, it's a "chromatogram", not a "spectrum".

With that nit-pick aside, the problem could be either with the column or with the pumping system. Since you have another instrument, the easiest diagnostic is to switch columns and see if the problem follows the column or stays with the instrument.

Another hint is to look at the dead time marker (sometimes called the "solvent front"). If the dead time and the retention times of all your analyte peaks changed *by the same percentage*, that strongly suggests a flow problem (leaks, bad seals, bad check valves, air bubbles in the pump head, etc.).

Good luck (and let us know what you find out, so we can all learn from it!)

By Alexander on Wednesday, July 7, 2004 - 07:00 pm:


One typical error people make monitoring reaction mixtures by HPLC is forgetting about compatibility of the reaction media and the column. Make sure you adjust the sample pH to neutral or acidic before injecting to RP column. Make sure other ingredients are not irreversibly sticking to the column.

By Dan on Thursday, July 8, 2004 - 07:05 am:

Hi guys,
Thanks for your help. Today the problem is a little different. I ran the same sample on 2 instruments this morning, including the one that was giving me problems yesterday ("my instrument"). My instrument produced a chromatogram (thanks Tom) that was identical to what I had seen in previous weeks. The other instrument was similar, but retention times were about 0.5 min shorter and it picked up a minor peak that my instrument did not.
The compound itself is compatible with the column and mobile phase and should not stick to the column. There are unidentified impurities in the reaction mixture, but I don't think they should cause a problem either.
Any suggestions?

By tom jupille on Thursday, July 8, 2004 - 09:16 am:

If "your instrument" is back to normal, that suggests yesterday's problems were causes by a flow problem (a trapped bubble would be my guess).

The instrument-to-instrument difference you're seeing could be due to a couple of things:
- the most likely is a slight difference in the selectivity of the two columns. In many cases, retention and peak spacing can be affected by very subtle differences column chemistry (including those brought about by the aging of the column).
- a second possibility (if your separation involves the use of a gradient) is a difference in the gradient delay volume between the two instruments. If your separation is isocratic (i.e., mobile phase composition remains constant during the run) then you can ignore this.

Quickest check is to swap columns between the two instruments and see what happens.

By Dan on Thursday, July 8, 2004 - 10:04 am:

Hi Tom,
I forgot to mention in the last update that I did try swapping the columns between the two instruments. The retention times were constant to the column. That is, I saw the same retention times with the columns as before, but on the different instruments. So, your assesment of a flow problem seems most likely for the original problem. The only question I have left is what could cause that type of problem? Should any maintenance be done to prevent it from happening again?

By tom jupille on Thursday, July 8, 2004 - 12:15 pm:

Since the problem went away, the most likely culprit is a bubble. General remedy is make sure you're degassing effectively. Also check the "sinker" frit on the inlet line to make sure it's not partially plugged, and check the inlet lines themselves to make sure they're not kinked (a flow restriction on the inlet side can cause enough of a pressure drop to persuade residual dissolved air to come out as bubbles.

When all is said and done "bubbles happen". If you're getting longer-than-usual retention times (and, like I said in an earlier post, if all the peaks are off by the same percentage), just re-prime the pump and run again.

Since you've pinned down your selectivity difference to the column, you can now look at what might be causing *that*. Are the columns from the same lot? Different ages? Different histories? All of those can contribute to selectivity differences. (if HPLC were easy, we'd all be unemployed!).

By Dan on Thursday, July 8, 2004 - 12:50 pm:

I checked the inlet lines for kinks. I can see a very minor kink, but it doesn't seem to me to be enough that it would reduce flow and drop the pressure. Am I wrong? I mean, can a slight bend cause a flow restriction?

Selectivity differences between the two columns I tried doesn't surprise me too much. The "other" instrument I used belongs to someone else, so it has a much different history. I belive both are from the same lot, but mine has been used more heavily.

By tom jupille on Friday, July 9, 2004 - 09:58 am:

sounds like your system is basically ok. Like I said in the earlier post, "bubbles happen". If the problem doesn't recur, don't worry about it. If it happens again, you may want to think about having your pumping system serviced.