Generation of Standard Curve and Quantification

[Medic1988]

Hi all!

I am hoping that someone can assist me in answering a few questions that are currently doing my head in. I'm using RPHPLC to quantify a total of 12 isoflavones in a variety of soy foods.

My first question is do I require a different standard curve for each of the 12 isoflavones? Or can I simply do 1 and apply the formula to all 12 for quantification?

Secondly, when calculating the quantity, do you need to take into account the injection volume? E.g. I have created a standard curve using concentrations of 0, 2, 4, 8 and 16 ug/ml of a standard. I have run each in triplicate, averaged the peak area and plotted a standard curve. When I identify my peaks from my food samples, I can insert the value of the peak area into the formula to quantify the isoflavone in question. But, do I need to take into account the injection volume? The process seems too simple and I feel like I'm overlooking something.

Sorry for the story of my life!

Any feedback would be much appreciated.

Cheers

[tom jupille]

do I require a different standard curve for each of the 12 isoflavones? Or can I simply do 1 and apply the formula to all 12 for quantification?

In principle, you need standards for each. In practice, if you can demonstrate (prove!) that all compounds have the same response factor, then you could get away with a single standard.

Secondly, when calculating the quantity, do you need to take into account the injection volume?

So long as the injection volume is constant (i.e. the same injection volume for your samples and your calibrators), you don't need to take it explicitly into account (volume effects cancel out). If the injection volumes will be different, then you should express your calibration as "mass on-column" (i.e., concentration times volume injected).

[lmh]

if you are looking at a family of similar compounds where only some are available as commercial standards, it would be reasonable in an academic environment to use the things that are available to calibrate those that aren't, with two provisos: firstly you need to make it clear what you've done, and that it carries an element of doubt. Secondly, it only makes sense if the calibrating and target analytes have very similar chromophores. For typical plant phenolics, it makes sense where the phenolic bit is the same, but they vary in their glycosylation, as the glycosyl groups won't have an absorbance in the range used to quantify the phenolic, and hopefully won't influence the phenolic's extinction coefficient too drastically.