Questioning the use of response factor for calculating conc.

[ccahalane]

Hello all,

I'm reading a method validation report wherein the amount of compound (let's call it M) is calculated by using a response factor. Compound M is extracted into 1 mL of extraction solvent from 1 mL of 1:1 Acetonitrile:water solution. This is a GCMS method and they are calculating the response factor (RF) by using the conc. of compound M (mg/L)/peak height. They then use the formula: peak height x RF x (mL extraction solvent/mg compound M) to calculate the amount of compound M in ppm.

I'm having a hard time accepting this as a valid way to calculate the amount of compound M. I think it assumes a 100% extraction efficiency and also seems a little circular. Shouldn't there have been an internal standard used to calculate the RF instead? Or use an external standard of known concentration of compound M made up in the extraction solvent to truly calculate the amount of M in a sample as well as the extraction efficiency?

Thanks for your time!

[Steve Reimer]

GC/MS usually uses an internal standard technique but it's not required. From what you have given us if the response factor has been determined from the calibration standards at varying concnetrations then the RF X peak height gives the concentration in mg/L. That is ppm in a mass per volume sense. Multiplying that by mL of solvent / mg of M doesn't make sense. If you wanted ppm in mass per mass that multipier would need to be one over the solvent density. Could that be it?

[Peter Apps]

The problem is not in the use of the response factor but in the equation at the end of the first paragraph - once you have mg/L from peak height x RF you multiply that by the volume to get the mass of analyte.

You might find it easier if you express all your quantities in proper units of mas or volume rather than using the ambiguous "ppm".

Peter

[ccahalane]

Thanks for the responses!

Peter - I didn't write this validation report. I wish the authors would have used proper units instead of ppm so I could have a better understanding of what they were calculating, hence the original post

[rb6banjo]

I do agree with you on the problems with the method. You rarely get 100% extraction with a method like this. I would like to see an internal standard that has similar solubility characteristics to compound M that "comes along for the ride" during the preparation.

In the method, do you first dissolve a solid sample into the ACN/Water mixture and then extract with the solvent for analysis? The units will work out if you divide by the size of the sample (in grams) instead of "mg of compound M".