Hi,

I have perfectly symmetric peak wich elutes on down slope baseline. I have USP tailing factor on my system suitability and this faills since the factor doesn't take the baseline drift in to account. Anobody had a same situation. Is there a symmetry factor available which takes this in to account?

Interesting question...

I don't know if this will help or not, but it may be worth investigating, just to see what sort of results you do get.

Why not subtract a blank gradient chromatogram from one with your sample peak, then run your USP tailling calculation on the resulting chromatogram? I would not suggest actually quantifying your peak using the subtracted chromatogram, as there are too many ways to get into trouble there, but if your gradient is nice and reproducible in the vicinity of your peak and there aren't any small excipient or gradient peaks in the neighborhood to annoy you, it might be worth a shot.

Whether this would be seen as acceptable by anyone is an open question, but it might help bolster your argument that you do really have a symmetrical peak.

Merely my 2 cents mind you...

Christopher

In principle, the tailing factor should be determined parallel to the base line. I do not know if the standard software is capable of doing this. If nothing else helps, you can do it manually...

Without some specifics of your method... why is the baseline sloping? -- can you not adjust your gradient conditions or eluent make-up to avoid/minimise this, eliminating the problems in determining the tailing factor.

Rereading my last message, I found it not to be clear:

If you have a drifting baseline, you can extrapolate the baseline under the peak. Now you drop a line from the tip of the peak to the sloping baseline. Next you draw a line parallel to the baseline at 10% of the height of the peak. You measure the width of the front of the peak and at the tail of the peak and do the calculation that USP requires. This is the correct USP tailing factor. It should not be a difficult task for software to deal with this.

Thank you. I use chemstation software. For USP tailing it drops a line from the peak top to the time axis (not to baseline as Uwe says). Then it draws the width line at 5% height parallel to the time axis (not to baseline!). If the peak is on down/up slope this method is totally wrong and it does not tell anything about peak symmetry. My peak of intrest is on valley. Sometimes it is on down slope and sometimes on up slope. I collected some symmetry data and found that only thing what USP tailing tells in my case is weather the peak is on down or up slope.

Is this a software bug? How does other softwares calculate USP tailing (millenium, chromeleon??)?

Uwe, do you have a reference, it would be really usefull? I haven't found any clearly stating how this should be done. It is a validated routine method and I wouldn't like to add too much time consuming and inaccurate manual calculations.

The principle of this is in my book "HPLC Columns" on pages 10 and 11. It does not deal with the USP tailing factor specifically, but with peak widths on a sloping baseline.
I did not yet check how different software versions are dealing with this situation.