Chromatography Forum

I have attempted Kaempferol assay.

Extraction/Diluting Solution: Methanol
Blank: Methanol

I made one stock standard and transferred 2 mL aliquots to 50, 25, 5, and 10 mL volumetric flasks (the 10 mL is the Bracketing Standard). The correlation coefficient is 0.99999 for the 50 mL, 25 mL, and 5 mL working standards. The Bracketing Standard at 10 mL is 87%. I repeated the aliquot dilutions to verify results and got a 93% recovery. I've already checked the integration of my peaks. I can only assume this is an issue with the Kaempferol, but it is strange that I would get a passing r^2 on my calibration curve yet a fail on the bracketing standard.

Any tips/suggestions?

If I understood correctly, the response of your highest concentration is lower than it should be. You might be simply out of the linear range of the detector.
I'd try halfing all the concentration by pipetting 1 mL instead of 2 mL in all flasks.

Hi Whelwig,

Agreed with HPLCAddict, this is the first thing I would try. The concentration range you're trying to cover is "relatively" broad (say, as compared to the API in a pharmaceutical finished product, 10 in this case, while 2 or 3 in a typical pharmaceutical HPLC assay)...it's worthwhile to halve all concentrations, check the slope of the calibration curve against what you have with your current dilution scheme and see how the second-highest concentration level responds.

Dually noted, thank you for the insight. I will look into this in further analyses.

Assuming that you are using a generic lc-method and there are no other problems with the lc-method or the implementation, I think I must disagree with the mentioned statements above.

Your concentration range is to narrow not too broad.
I agree to HPLCaddict, that you may out of the linear range of the detector. But simply take 50 % of the concentration would not help you. My Idea is that you are widely above the linear range of your detector. That means you concentration is simply too high.
(This would also explain the excellent correlation factor)

I would suggest you to make some initial experiments with dilution series over a real broad concentration-range. Accurate work is not necessary at this moment; make some 1:100 dilutions e.g. 10%; 0.1%, 0.001% and 0.00001%.
Start with the lowest concentration and observe if there is a negligible signal.

Please be so kind and add some basic information, like complete method-description, used instrument and chromatograms.

Hi whelwig,

Klaus has a point; breadth of range inclusive. Looked up a nice paper, Journal of Pharmaceutical and Biomedical Analysis, 48 (2008) 629-635. Author reports a nice method for which linearity of kaempferol was demonstrated as 0.209 - 16.712 micrograms per mL. Please, see how this correlates with what you're trying to do.

Best Wishes!