A number of our analytes are positively charged and v.
hydrophilic. I've used TFA and HFBA
(heptafluorobuytric acid) at 0.1% and 0.05%, in H2O
and MeCN (gradient method) to improve retention of
these compounds.

Even after removing the column and flushing the
system (a Finnigan LCQ Advantage ion trap) with 50:50
MeCN:H2O @ 1 mL/min for 3 hours, the HFBA was still
v. visible (a series of ions with intensities between
1x10^6 and 2x10^6 in ES- spaced at intervals of
236 amu--the MW of the HFBA ion), and I can get no
reponse at the expected m/z for flow injections of a
known compound (5-fluorouracil) under conditions that
have worked previously. It's difficult for me to believe
the the HFBA is still around at levels significant enough
to affect ionization after such a long flush. But, then
again, its ions remain v. visible.

Does anyone have experience with this? I see the
perflorinated acids used fairly frequently in the LC-MS
literature and have read about their ability to reduce
ES+ response and obliterate ES- response while being
pumped through the system, but not about their
"hanging around" so tenaciously.

Thanks, in advance, for your response. I hope
everyone had a safe and relaxing holiday.

I no longer use perflorinated acids in LCMS for this reason. I have had a problem with TFA, for example, sticking around for months, even when a different column was used. The affect depends on the analyte, so you may think the problem is gone, then you will find it again with another method. The only suggestion I have is to change all of the tubing, fittings, probe, etc that you can and flush with more water. If someone knows of a better way to get rid of the problem, please let me know.

I do not have experience with this, but to throw everything away does not make any sense to me. My first bet would be that TFA and HFBA will stick to the teflon tubing that is used to connect the HPLC system to the solvent bottles. I would either flush all the teflon lines with an organic solvent, or I would replace these lines with fresh tubing to see if the problem goes away or get better. Of course, there are plenty of things inside your HPLC as well that like to adsorb (or maybe even absorb) fluorinated stuff. But the chances are that the whole problem is in the previous water line.

PS: let me know how you made out - I am curious - the last anonymous

What is the difference between absorb and adsorb?
Please advise. Thanks!

Adsorb is on the surface and absorb is inside the material.

Jeff, this persistence of HFBA on the column could just be the solution for a problem we have with a TSK Gel Super SW 3000 column (exclusion...). This had a permanently lower performance (sharpness of protein peaks) after using TFA and HFBA in the mobile phase, even after having been run for many weeks without TFA, etc., in the mobile phases.

What column did you use?

First of all, thank you for all of your responses. I also
polled some friends who work with LC-MS and have
read a number of stories of people's problems after
using these acids. Let me tell you what I know at the
moment.

1. Cleaning the ion source thoroughly removed most of
the HFBA masses from my spectra. Removing the ion
source completly and immersing it in H2O followed by a
MeOH rinse reduced the HFBA masses significantly
(spraying the ion source with MeOH:H2O while
attached was not sufficient). Reattachment to the LC
brought some of the masses back (though at lower
intensities than previously observed).

2. Unfortunately, I had a confounding problem that may
or may not be related to the TFA/HFBA. My ES-
sensitivity and mass accuracy (a problem only in ES-
mode) were recovered only after cleaning the ion optics
(2 octapoles in my case). They appeared to be clean,
but the service rep came and, after we cleaned the ion
source and replaced the ion transfer capillary, he
suggested that we give it a shot. It worked. Whether
this was related to the HFBA I don't know.

3, I was using a Synergi Hydro-RP column. And, the
HFBA seems to have changed the retention
characteristics of the column. Now, I don't get good
peak shape without an ion pairing agent present (this
is after a number of hours flushing the column with
50:50 MeCN:H2O).

For the moment, I'm trying to avoid fluorinated acids in
my system. I've received responses from a number of
people who have come to the same conclusion. Others
have suggested it's "throwing out the baby with the
bathwater" and that there are likely ways to flush the
acids form the system. That's probably true, but since
my company is small and uses the same system for
trace analyis and routine analysis of synthesis
products, I'm being conservative for the moment. I'm
going to give HILIC a shot, as I am unable to get decent
retention (0.251.25) with acetate and formate
mobile phases.