Reduced recovery when analyzying Chlorhexidine Gluconate

[cody84]

I'm using an Alliance with a stainless steel needle and getting carryover.

Test your wash waste if it is separated from other waste!

Are the detectors both PDA or one PDA one dual/single?

New column, vs old column? Try backflushing both and see.

I don't think CHG would have any interactions with a normal and clean system with stainless steel etc. Not that I've seen anyways I've only tested it on an Alliance and a Varian Prostar.

If you're thinking needle and sample loop maybe check injection accuracy on both systems.

If possibly purge injector and do single injections

[Labmonkey]

I was using a Waters alliance and H-class and replicated the exact same results, I have not tried testing the wash waste but that is a good idea, thank you for that one.

I have analysed using a PDA and TUV detectors with no difference, and tested using old and new columns of the same and different batches. Precision of the instruments is not the problem as the systems are routinely checked for injection precision, and the results are all reproducible and fine for all other methods.

I didn't actually think that chlorhexidine would stick to stainless but i have read in articles that it can, and the platinum needle is one of the things shimadzu promotes for the prominance to help reduce carryover and sticking.

[cody84]

That's weird, because we haven't seen that happening here and we have stainless.

We're using the USP method though, maybe that's why.

[cody84]

Actually, I don't know if the TEA would even make it through the injection unit because its only in the buffer for the USP method...and maybe our carryover is the CHG sticking in there and slowly being released...??
Hmm now you have me thinking more, thanks!

[Labmonkey]

Had this not been a method transfer i doubt we would have actually noticed that there was an issue with the chlorhexidine in the first place.

[cody84]

Don't you do accuracy and range? (Recovery of analyte vs. theoretical concentration)

[Labmonkey]

It is not requirement of this company to do accuracy and range during a method transfer as the method was deemed sufficiently validated by QC management.

Had we performed it i believe they both would have failed.