Over- or under-reporting peak area?

[Ste_C]

Hi,

We are running a UPLC method where the flow rate changes from 0.16ml/min to 0.26ml/min after elution of our main peak of interest. There are several excipient / impurity peaks observed during the run before and after the main peak. We have modified our method to remove the flow rate change, now we have a concern about peak areas reported after the flow rate change. When results are compared between methods with flow rate change and method without flow rate change (constant 0.16ml/min) there is a significant difference of about 40% increase in peak area for the peaks after the main peak when the flow rate change is removed.
This is a gradient method using Acetonitrile and Water as mobile phase. The gradient was modified to correct for changes in retention times when the flow rate is changed.

Why would there be such a difference in peak area values when the flow rate is changed? Have we been under-reporting our %impurity values as a result?

[tristanewalters]

Hey

Part of how "large" your peak is has to do with how fast it is moving through the detector cell. At a higher flow rate, your molecules are traveling faster through the detector cell, at a lower flow rate it is the opposite. If your molecule is traveling slowly, chances are it will caught by the detector at more sampling points than if it was going faster.

So, for example if you had the same LC method but one method had a 1ml/min flow rate and the other had a 2ml/min flow rate, the molecules passing by the detector going 1ml/min would be "seen" approx. twice as many times by the detector as those molecules going 2ml/min. This would lead to approx. 2x higher peak area in the 1mL/min run than in the 2ml/min run

[lmh]

... ooh, so you mean the units of peak area are AbsorbanceUnits.Minutes, and actually, it would be better if chromatography systems gave peak area in units of AbsorbanceUnits.mL?

[tristanewalters]

I think that the peak area is a function of the detectors absorbance vs the time the peak is eluting. The peak area of the analyte of interest has many variables such as flow rate, temperature of mobile phase, composition of mobile phase, etc. I would assume that if you kept all parameters the same on a method, but doubled the flow rate, the peak area would be cut in half.

[DR]

^ is approximately true...

Instruct the author of the version of the method that incorporates a flow rate change to learn what a Van Deemter plot is and what it means. Unless you've stopped integrating peaks of interest, flow should remain constant. Validating a method with responses recorded in terms of area/mL would be (to grossly underestimate it) "difficult".

Further instruct the method author that changing retention times are not to be cured by playing with the pump settings, but by correcting hardware problems or letting column chemistry come to a state of equilibrium prior to injecting anything important.

[HPLCaddict]

I think that the peak area is a function of the detectors absorbance vs the time the peak is eluting. The peak area of the analyte of interest has many variables such as flow rate, temperature of mobile phase, composition of mobile phase, etc. I would assume that if you kept all parameters the same on a method, but doubled the flow rate, the peak area would be cut in half.

You're right with the flow-rate, but temperature and mobile-phase composition will not influence peak areas.

[tristanewalters]

Are you sure about that? As mobile phase %B composition and column temperature (i said MP temperature but this is what I was referring to) affects retention and, assuming isocratic elution, retention time changes peak width (I assumed area too but this could be my mistake) which could lead to different peak areas.

[HPLCaddict]

Yes, you're right that a different retention time in isocratic elution will lead to different peak width, but the peak height will also change . If you're doing decent chromatography (i.e. no coeluting peaks), peak area will remain constant.
For UV detection, peak area should be a function of sample concentration (obviously), injection volume, flow-rate and detection wavelength. For ionizable compounds, there CAN be an additional pH-dependance, if the absorption coefficients of ionized and neutral form are different. Hope I didn't miss anything[:)].