Use of Anti oxidant in Mobile Phase

[adam]

Hello

I would like to ask if anyone has experience with adding antioxidant to the mobile phase, or to the diluent.

We are currently running an analysis, and one of our analytes is tretinoin, which is somewhat susceptible to oxidation. We find that we get good results most of the time. But maybe one run out of 20 (or something like that) we will get a low tretinoin result. For example we'll get a result of 88 or 89% label claim, when all the other results were close to 100%.

We know these results are not real, and are somehow due to the analytical method. My guess is that trace oxidants somewhere in the system are causing this. So I am considering adding an antioxidant (such as BHT) to the diluent or mobile phase.

I've never done this before and I wanted to ask if anyone has any experience with this, or any suggestions to share.

Much Thanks
Adam

[tom jupille]

Adding it to the mobile phase makes sense, but you have to verify that there is no interference from the BHT itself or it's breakdown products. I would try that before adding it to the mobile phase; the residence time of tretinoin in the mobile phase during the chromatography is probably a lot shorter than the residence time in the autosampler vial.

[DJ]

How do you know it is not real?

What is your sample dissolved in?

[JMB]

One out of 20 results is low (Never high ?)---I would suspect a bad injection.

What result does a duplicate of the "bad" vial give ?

JMB

[Gerhard Kratz]

Before I would add something to the mobile phase I would try cooling the vials in the autosampler if that is possible.
A little bit SDS in the MP - no problem, BUT it is a change in your method!

[carlo.annaratone]

Why don't you add it in the sample vial?

[adam]

Thanks all

Several of you commented that adding it to the diluent would be preferrable, and the thing to try first (due to higher residence time in the diluent vs mobile phase, less detection interference, etc). I think you're right and so we will try that first.

Gerhard: I'm curious to ask, you mentioned SDS (which I'm assuming to be sodium dodecyl sulfate). I'm just wondering why you suggested that as an additive. I don't think of SDS as having antioxidant properties. I was just curious to understand your thought process.

Thanks all for the feedback

[danko]

One out of 20 results is low
For example we'll get a result of 88 or 89% label claim, when all the other results were close to 100%.

Adam, in this one out of 20 injections where 11 – 12% is missing – do you see a new peak that you don’t see in the 100% injections? Or maybe a higher injection peak?
And the million $ question: Have you asked yourself why this “oxidation” only occurs sometimes – the diluent and the mobile phase is the same for all samples/injections?

Best Regrads

[DJ]

Why oxidation?

[bunnahabhain]

Danko is right, why do you see this only in one out of 20 injections? Are you at university or in a commercial lab? I remember back in my university days we re-used sample vials after more-or-less thorough cleaning. Actually, not only vials but also caps and micro-inserts were cleaned and re-used. Having one out of 20 vials contaminated with some sort of catalyst killing your analytes is well possible then. Solution: Convince your boss that vials REALLY are single use material.
If you are using new vials, and only air oxygen was the problem, you would see a time-dependent decrease of analyte rather than outliers. Ensure that all your caps are closed tightly and retry.
Another experiment to perform would be: Take two sample vials with identical content. Store one of them under Argon atmosphere and bubble air through the other (not too much bubbling, it will remove water and thus increase concentration of your analyte. Maybe weigh the whole vial before bubbling and add water or clean solvent to get to the initial weight after the bubbling). Then analyse both vials and compare.

Good luck,
Jörg