Related Impurities analysis: RRT of 2.22 = RRT of 2.23???

[vliu]

I am performing a validation on the related impurities. I have just ran my day 30 samples and is trying to compare to my day 0 results to determine the stability of the samples.

The criteria requires to measure and report impurities if it's >0.05% of the main area peak. The unknown impurities will be denotes by their relative retention time (RRT), my question is - when denoting a new impurity by its RRT, how much tolerance should I allow the RRT to shift ie. b/n 2 runs if the RRT shifts from 2.22 to 2.23 or 2.24, should I still classified them as the same impurity? To make the whole situation more complicated, I was asked to track ALL impurities for Day 30 sample analysis even if it's <0.05% of the main peak with s/n <10, at this low response it's really hard to determine if RRT =2.22 from the first run is the same when compare to the second run with a RRT= 2.23 or 2.34.

Any advice or suggestion would be greatly appreciated. Thank you in advance.

[tom jupille]

As with so many things, the answer is "that depends": how many impurities did you have in the earlier runs, and were there any in the same region of the chromatogram as this one?

It's really a question of specificity; the appropriate "window" is halfway to the next peak in each direction. This assumes, of course, that no *additional* peaks show up in that retention window.

[sassman]

We typically give it a + or - of 0.02.