how to remove the major API to detect the degradants

[ym3142]

Hi, everyone;
To detect possible degradation, we are looking into all the tiny peaks in the LC until they are real noise. But whatever the condition is, we have a huge API such as 800mg of guaifenesin out of 1000mg tablet. Does any one have the experience of removing these API before detecting the degradants in LC(MS), where we can rule out the possibility of degradants overlapping under large API peaks? Thanks in advance and have a great holiday season,

[SIELC_Tech]

Excel,

Guaifenesin already have two impurities coming from the production (two isomers of Guaiacol). We have a method for determination of these impurities in Cough Medication formulation (also contains acetaminophen and other ingredients). Resolution between peeks is around 2.5-4 and it is on the level of 0.1%.
Let us know if you are interested (it is not on our website yet).

Regards,

Vlad

[DR]

We haven't worked with that stuff, but we generally do not remove API. If the API peak is that huge and the impurities' extinction coefficients are anywhere close to that of the API, we just characterize down to a LOQ spec determined in validation and we don't report any unknowns that are (typically) <0.05% vs. API. If named peaks are there but very small, we report them as <0.05%.

For oddball impurities and other related substances that have much lower absorbances, we would be inclined to run them via a different method that resolves them from the API completely. Then it wouldn't matter if the API went off scale as we would be comparing the RS to a nominal API concentration.

[tom jupille]

ym3142, I was a little unclear on how to interpret your last sentence. Are you looking for a way of getting rid of the API so you can see any small degradants in that region? Or are you looking to get rid of the API (so as not to swamp your MS) and assuming there are no degradants in that region?

If it's the latter, you can set up a simple switching valve to divert flow to waste while the API is eluting.

If it's the former, I can't think of anything that would get rid of the API and not also risk getting rid of close degradants. What you could do is to have two "orthogonal" methods (so that the close degradants would be different in each) and use the diversion valve trick on each, but this is compounding your method development problems!

[ym3142]

Vlad, DR and Tom,
thank you very much to take time for looking into my problem.
And Tom, For

Are you looking for a way of getting rid of the API so you can see any small degradants in that region?

the answer is Yes.
And what is

use the diversion valve trick on each

?

Vlad, I am really looking into those more than known impurities from all the API's. Thanks though.

[SIELC_Tech]

Diversion valve might be tricky in your case. Here is the newsletter on this approach (I think that what Tom had in mind):

http://www.sielc.com/pdf/SIELC_September_2004.pdf

In your case your are tryimg to divert you main product.

We never tried to analyze degradation products for Guaiphenesin as we had a task to develop method for current impurities, but don't see problem in analyzing degradation products with the same method (I will send you method later today). Have you tried GC for this one (derivatization or straight)?
Vlad

[ym3142]

Vlad,
Thank you again.
I tried GC, which faied to detect the degradation. Though it was done by an outsourceing analytcal lab, the chromatogram was not well resolved, either the column was overloaded or many noise/column bleeding were picked.
I will check you reference out.

[SIELC_Tech]

Excel,

I emailed your two files (gmail account)-one on guaifenesin and another one on the decomposition study (good example for Primesep columns))

Regards,

Vlad