Relative Response Factor - Limits of Use

[Rob Burgess]

I have just calculated a HPLC relative response factors for a known degradant that gives a value of 0.05. (NB. Calculated as the Response for Imp. / Response for Main Peak Analyte).

I have read some internal guidelines that suggests that for RRF’s less than 0.2 (and greater than 5), the method should not be used “as isâ€

[tom jupille]

I can speculate:

a) presumably you're looking at λmax of the analyte. If the RRF for the impurity is that low, the absorbance spectrum of the impurity is quite different, which implies you are far from λmax of the impurity. Trying to do UV on the slope of an absorbance band is always problematic, because response can shift dramatically with conditions.

b) even with comparable response factors, you often have to overload the daylights out of the main peak in order to get to LLOQ for the impurity. A low RRF makes it that much harder.

[Rob Burgess]

Cheers for you response Tom... again much appreciated.

Can I just add that I had the ratio wrong in my original mail: The RRF value of 0.05 was calculated from: Slope of main API peak / Slope of impurity at a wavelength of 276 nm.

Also can I ask (for anyone out there) would this RRF still be applicable to other chromatographic conditions were I not to change the wavelength of analysis but changed some other parameter i.e. if I changed the gradient conditions of changed mobile phase entirely. If I were to make these types of chnages would that warrant re-calculating the RRF again. I'm presuming I would need to recalculate.

[DR]

If a standard of the impurity is available, you should go ahead and check the response ratios when you change conditions. I would be surprised if the ratio changed...

When you have a significant change in absorbance due to formation of a ring, additional double bond or whatnot, leading to a much higher RRF, I guess you have to worry about linearity - but that shouldn't happen until RRFs get well above 5 (unless you form significant quantities of said compound at which point RRFs cease to be anywhere near the top of the list of problems you're facing )

[tom jupille]

I would be surprised if the ratio changed...

I would be surprised, but not shocked. As a trivial example, something as simple as TFA forms a charge-transfer complex with acetonitrile; as a consequence, the absorbance spectrum of TFA changes with ACN content.

Granted, that sort of thing is the exception rather than the rule