I am in the pharmaceutical industry and I have a drug substance with RT of 8 minutes. When viewed full scale, the peak appears to start at 7.6 minutes and returns to baseline at 8.5 minutes. I am using Waters Millenium software and it calculates a USP tailing factor of 1.1. The problem is that when the tail of the peak is examined closely by zooming in, it does not return to baseline until about 9.3 minutes. The peak has a maximum absorbance at 8 minutes of 0.357 AU and at 8.5 minutes it is 0.0004 AU. When the chromatogram is printed with the peak in full scale, integrating to 9.3 looks incorrect because it looks like the peak ends at 8.5 minutes. What is the correct or proper way to integrate, stopping at 8.5 or 9.3 minutes?

Thanks in advance for any advice or generally accepted practices.

Steve,

Why don't you measure the peak areas which result after stopping at 8.5 mins or 9.3 mins, then we can get some idea of the potential error that you may experience.

David

David

The area after integrating at 9.3 minutes is 0.187% higher than at 8.6 minutes. (5285349 vs 5275483)The specification for this assay is 98.5% to 101.5%.

Steve,

This does not sound like a giant error to me. I would think that in theory the larger value is the better one; however, you may be integrating noise in the larger measurement.

You are lucky that your USP tailing factor is only 1.1. I guess you would have much more serious problems with peaks showing more tailing.

I don't think there is an answer to your question. Perhaps you should also investigate the reproducibility of peak areas with different integration parameters (e.g. integrating at 9.3 mins or 8.6 mins ) to see if this should be a factor in your decision.

David

David,

Thanks for your thoughts on this subject. %RSD is 0.23% for both. As far as error goes, the effect on quantitation is negligible since the std and sample are prepared at identical concentrations so differences in integration affects both the same. Like you said, in theory the larger value is the better one, but if accuracy in quantitation is a major factor and the error in integration is very small then I guess it should be okay to use the 8.6 min integration end point.

How is your calibration standard integrated? In doubtful cases I look that both calibration and sample are integrated similarily. This should reduce any errors.

K.H.W.

This is an API raw material assay so the std and sample are essentially identical in size, shape and integration. That is why I said that error should be neglible.

I'm having an exam tomorrow and i was wondering if anyone can clarify to me the following:What impact does the efficiency, selectivity, capacity and tailing factor have on chromatography ( Hoe does it affect the peak, retention time, resolution, quantitation,,,calibration)
If anybody can help me i'll be greatful!

The time to ask this sort of question is during/after the lecture, *not* the day before the exam. However, the answers can be found at:

http://www.lcresources.com/courses/getting_started/