Estimating response factors (m)

[domino1]

Hi all.

I have been hearing about estimating relative response factors by "serial dilution" for peaks where you do not have an authentic. (I am guessing that you are comparing slopes).

Can someone shed some light on exactly how this is done? I just can't seem to get my head around it when I don't know my starting concentration.

Sorry to be so clueless - but I never claimed to be a math whiz!

Thanks,

Domino

[ntruong]

Can someone shed some light on exactly how this is done

Check the working concentration of the method that you are interested in setting up RRF values. If the working concentration of the main compound is 1.0 mg/ml, for instance, you would treat this as 100% then prepare a serial solutions of 50%, 75%, 100%,125% and 150% to determine the slope(A). Prepare a serial dilutions of any known impurity to that of main compound from 50% of 0.1% per ICH guideline (or limit) to 120% of limit to determine slope (B). RRf = slope (B)/(A).
Good luck,

[domino1]

Thank you.

My only source of the impurity is in the degraded sample iteself (since I do not have an authentic).

If I do serially dilute this sample, I can plot the main peak area versus concentraiton since I can quantitate it against a standard. However, I can't plot the area vs conc. for the impurity since I don't know how much is in the sample to begin with. I can't use the main peak because I suspect it has a different extinction coefficient based on my lack of mass balance for the degraded sample.

If you could provide some hypothetical numbers to show me it would be greatly appreciated. Again, I am sorry that I can't seem to get my head around this one! I do appreciate your time!

Domino

[ntruong]

My only source of the impurity is in the degraded sample iteself (since I do not have an authentic).

Since you cannot synthesize or identify this impurity, treat it as standard. I mean you have to assume the absorbance of this impurity is the same as standard. Therefore, the RRF for this impurity is 1 with respect to known standard. I do it all the time and this is acceptable practice

If I do serially dilute this sample, I can plot the main peak area versus concentraiton since I can quantitate it against a standard. However, I can't plot the area vs conc. for the impurity since I don't know how much is in the sample to begin with. I can't use the main peak because I suspect it has a different extinction coefficient based on my lack of mass balance for the degraded sample

Make a serial dilution of your standard as you would if it were impurity. I don't know about the working concentration of your standard, but I would make it very concentrate in order to show the impurity. Let assume your working standard is at 1mg/ml (100%), you would typically prepare linearity solutions at 120%, 100%, 10%, 1%, 0.1%, 0.05%, 0.02% and 0.01% (you have to go down this low since the ICH guideline for impurity is 0.1%?I have to look it up ). You also need to determine LOD and LOQ of this standard to show that you are able to detect at a very low level.
Good luck,

[domino1]

The only problem is that I don't think the RRF is 1. That's why I need to calcualte it. I lose total area counts when I degrade the sample and I suspect it is due to extinction coefficient differences.

It is possible that the compound could be degrading to something that does not have a chromaphore. The extinction coefficient difference would just be an easier fix.

Thanks again,

Domino

[Rob Burgess]

I think you may have encountered one of the most difficult problems to overcome (i'm dealing with a simialr problem right now).

Depending on your urgency to get this done I can suggest a few things:

1) identify the unknown - MS/DAD/ NMR etc.
2) get someone to synthesis the now known impurity for you
3) failing that try prep-LC to get enough material to allow measurement of RRF's (also try combining with SPE to pre-concentrate)
4) you may also be able to mass directed prep. (its a buzz word going around at the moment - not totaly sure what this is - any offers on this one?)

[ntruong]

I would concur with Rob. He read my mind

[domino1]

Sigh. That is what I am afraid of.

It is just that I heard this comment twice about comparing the slopes of serially diluted samples to determine RRFs so I thougth that there must be something to it.

You CAN use the slopes if your main peak degrades into a single impurity of an unknown extinction coefficient and you plot area vs time (so levels of degradation). However, that is not the "serial dilution" experiment that I have heard of. Plus, things never degrade to just one species.

Thanks again. I have a meeting tomorrow that hopefully will shed some light. If I find out anything of interest I will let you know.

Domino.

[Uwe Neue]

Mass directed prep is the common solution to such problems. You do preparative chromatography with a mass spectrometer attached to the prep machine. It will tell you, where the stuff elutes that you need to collect, or conversely, it tells you the masses (and spectra) of the peaks that you are collecting.

Since the equipment is expensive, it is useful only under circumstances where such problems need to solved routinely.