Page 1 of 1

HPAEC PAD

Posted: Fri Aug 25, 2017 8:22 pm
by Harry64
Hi all ,

I 'm working with HPAEC PAD ... Unstable results, bad repeatability ..Sequence of 20 samples ... 18 OK , 2 NOK.

Calibration in diff. matrix OK ...using Disposable Gold Electrodes ...

What could be the reason for the unstable results ...

many thx for an answer

Re: HPAEC PAD

Posted: Sat Aug 26, 2017 2:23 pm
by mattmullaney
Hi Harry64,

What is the matrix--more or less? I've used this detection for fermentation broths, at the cusp of the release of the disposable electrodes, so I'm used to using MS-grade sandpaper to clean them up...or aluminum oxide paste. I found that I could get 50 - 60 injections before a change in response.

Please, let me know what you have experienced and think in more detail.

Re: HPAEC PAD

Posted: Sun Aug 27, 2017 12:19 pm
by Harry64
Hi Matt ,
Matrix is watery ( diluted ) , beers related products ... ...detection charbo's ...
You are not using disposable gold electrodes ... just a thick fix solid gold electrode ...and even then you have to clean the electrode after 50 samples ...
Do you also have to dilute your samples ?

rgrds

Re: HPAEC PAD

Posted: Sun Aug 27, 2017 9:09 pm
by mattmullaney
Hi Again, Harry64,

Your matrix is fairly similar to mine (fermentation broths for what is known now as Gardasil valents), so that isn't too much of a problem. Generally I did not have to make dilutions, but at some times, never more than one part to five parts or so.

Yes, though the electrode(s) in my case were of the older DX-500 variety before the disposable electrodes caught on, I still had to do the polishing every 50 - 60 injections...the eluent was fairly concentrated NaOH (also at the cusp of the eluent generator)...and it really is what is happening at the surface of the gold that is what counts. Lots of maintenance, yes. The performance while the electrode was working, though, was outstanding, to be fair.

This was back in 2001 - 2005 while I was at Merck. Also did some Pt electrode work for alcohols using IPAD. That was a little tougher, but also worked pretty well...made more sense though to simply use GC-FID in the long run.