Hello,
We have been having contamination issues with our single quad LC-MS system (Waters Acquity Arc QSMR, Sample Manager FTNR, 2998 PDA, QDa). In particular, we observe an omnipresent species with m/z 100 across the entire chromatogram, m/z 271 and a series with m/z 295, 297, 299 (as well as 293, 296, 298 and 300, see pic) when the aqueous phase is between 100 and 20%, after that it seems to disappear. I should mention we are running either 10 minutes or 30 minutes gradients (5-90% ACN, only FA as additive for both MPs). I have tried to isolate the problem to either LC or MS, with inconclusive results. We have performed a thorough LC cleaning twice (washed with acidic mix, ammonia and IPA, then water @70C overnight, all LC-MS–grade from Biosolve), as well as MS source and ion guide cleaning (done by Waters engineer), but the problem is still there.
The strangest thing though, is that after the first round of LC cleaning the contaminants were gone (or below noise level) for about two consecutive days (see pics of spectra), and then all reappeared on day 3 (except for 271) when a new run was started, only difference being that I prepared fresh mobile phase (with commercial LC-MS–grade water and LC-MS–grade formic acid). After that I tried the cleaning procedure again, but this time with no success, although the general background signal did get much better.
We even installed a carbon filter between our (newly installed and checked) N2 tap and the MS inlet. At this point I do not know what else to try, Waters keeps suggesting that it's an LC problem, and they advice yet again another cleaning procedure (with 30% phosphoric acid), but I am a bit skeptical since we have tried 3 out of 4 of their advised cleaning mixes and protocols already.
Did anyone maybe encounter these same contaminant species? Or does anyone have suggestions as to what they might be and how to get rid of them? Since this is just a single quad, it's hard to say exactly from the m/z, but if I scan some of the " common contaminants" lists some of these masses match with either Triton or polymers (PPG or PEG), although I do not see the typical patterns for the latter. Any suggestion or tip would be greatly appreciated :)
Thank you
I would start with anything that is being washed in the lab, for instance the bottles used for the mobile phase. If these go into general lab cleaning the issue could be surfactants from the detergent used in cleaning the glassware. I would then start to sub some other buffer for the FA - and organic - Like methanol and ammonium acetate to try and narrow down where the contamination is coming from. Maybe also try running negative electrospray - is the same contamination seen in neg mode? Does it give a higher or lower response? You may be unknowingly getting contamination from the strong wash / weak wash solvents. Try setting volume of injection to zero (sequence table) and make a number of injections - do you still see the contamination? If you have an infusion pump/syringe system to deliver tuning solution - sometimes that rotor seal can get scored and slowly leak infusion compound - I would do a thorough purge of that while you are at it, then infuse 50/50 Acetonitrile/Water/0.1% Formic - do you see the contamination. Do this for a couple purge cycles ... do the peaks get smaller as the infusion progresses? This could also narrow it down - I would suspect
1.) Bottles (check every one including strong and weak wash)
2.) Mobile phase
3.) Formic acid
4.) infusion rotor scored
Pull all the bottles off - rinse with water, then methanol
good luck!
Thank you for your answer Peter!
We have glassware that is specifically for LC-MS applications and we never wash it with anything else than LC-MS–grade solvents and water. I also tried to replace ACN with MeOH and saw the same contaminants, as well as if I only use ACN/water without FA. Tried water from different sources as well, even bottled commercially, and again no differences. So I thought I could exclude the solvents.
I forgot to mention that in ESI- we have a very clean baseline, especially after the rounds of cleaning, so these contaminants only appear in positive mode.
I also tried zero volume injections (60x, 5min gradients) and manual infusions (with water/MeOH, water/ACN, both with and without FA) and always saw these m/z species at constant intensities. During gradient runs they do not elute as peaks but rather are present across the entire chromatograms. The only difference is that as the percentage of organic MP goes up, then the signal intensity goes down for these specific masses, but I don't know what to make of that, given that the water per se should be fine.
We do have a divert valve and that is the only thing that remains on your list that we haven't thoroughly cleaned, because with the cleaning procedures suggested by Waters usually you have that set to waste so only that path gets cleaned. I could try cleaning the MS path manually by repeated infusions strong solvents, but is it possible that something wrong in the divert valve is causing such a resilient contamination that "never" washes away?
The puzzling this is also that we had this machine for 1 year without issues, then we moved it to a new location and these m/z species appeared. I even started to think that maybe the cleaning staff uses floor cleaning machines in our analytical chemistry lab (although they have been instructed not to); could those be spraying detergents in the air maybe and they get in bottle/MS?
Best,
Riccardo
I once had an issue with the HVAC system - one of the air handling fan belts got shredded, and it rained fine particulates all over my solvent tray. I've covered my solvents with aluminum foil ever since.
Yes that divert valve rotor seal could be scored - I've had that happen (I have Waters TQS micro) - best to purge it with methanol just to rule it out. You can also just infuse and see if your contamination peaks get higher when you switch the divert valve to infusion.
Have you tried running without the column installed to eliminate that possibility?
This sounds like a nightmare -
Yeah it's definitely not fun to deal with 😅
Yes, I did run without column with again no differences. And I also thought about things falling off into the solvent bottles (we have this extraction arm on top of the equipment that looks a bit old and crumbly), but I just bought proper bottle lids that have no openings, so that is ruled out as well (from now on at least).
Thanks for the tips about the divert valve, I am going to investigate that and do some tests and see if I can pinpoint the source.
Best,
Riccardo
- Badges
- Science method