Hi Florian,

the link below might give you some more info as to the likely origin of your contamination.

Frequent clogging would suggest something in your samples, or perhaps buffers. I'd be surprised if it originated from your buffers if they are MS grade. I assume you spin or filter your samples prior to loading? Are they likely to precipitate for any reason prior to loading? It's obviously hard to say if the contamination is related to the clogging, they may well be separate issues.

Good luck.

http://www.maconda.bham.ac.uk/contaminant.php?id=CON00051

This problem is due to Acetonirile and it can be resolved with slow gradient. I observed that when sharp gradient has been used then this peak intensity will increase. 

Hello Peter,

thanks. I was already searching into that direction.

The problem is that the intensity is really different within replicates produced at the same time.

Any more suggestions.

Thanks,

Florian

Hi Florian

We saw this problem with a 453 peak for a while some years back and traced it to the syringe filters we were using to clean up the sample extracts. Off hand I cannot remember the brand of filter, but we do use Sartorius and  Pall filters often but also others from local suppliers so it could have been from any of these. We changed to sample clarification using centrifugation at 16000 g instead of filtering as this resulted in less sample loss and avoided the problem of the contaminating peak. We found that after a while the peak would even appear in blank runs implying that the contaminant was showing retention in the column followed by slow leaching out during the analyses. This can be tested by running a blank run with the column removed and if the contaminant is gone then the column would need to be cleaned up using lipophilic solvents for an extended run time (just don't elute into your MS system but to waste).

Another way around the problem would be to pre-rinse the filters with your solvent to remove the contaminant before actually filtering the sample itself.

Good luck

If MS/MS on 453 precursor results in fragments 227, then MS3 on 227 leads to m/z 199 , then you do have the cyclic oligomer of polyamide 66 from nylon.

Maybe this will be helpful too http://www.waters.com/webassets/cms/support/docs/bkgrnd_ion_mstr_list.pdf

To determine the source of contamination, you could do direct infusion of solvents (bypassing the LC), then connect the LC to MS directly from the pump (bypassing column and pre-column trap).

Hi Florian,

the varying intensity does complicate things for you. Is there any pattern to it? ie. once every 3 samples etc? Is it always the first injection that has the highest intensity, or does it build up? You might be able to track it down that way.

Sometimes you are better off changing everything. On the good side it may well get rid of the problem and it's faster than changing things systematically. On the bad side, you don't know what caused the problem in the first place.

Peter

Hi Florian,

nominal mass 453 (m/z) could be 2 contaminants:

453.20966 [A9B+K]+ [C2H4O]nH2O PEG Polyethylene glycol, ubiquitous polyether

453.34353 [M+H]+ C24H44N4O4 nylon Cyclic oligomer of polyamide 66, (adipic acid/hexylmethylene diamine condensation)

According to Elena you could distinguish by MSn experiment or by High Resolution MS with accurate mass determination.

As the PEG is formed with K+ Ions, I actually prefer the hypothesis of the cyclic Polyamide.

Kind regards and good luck

Guido

dear floria

if the neutro loss of 113 can be observed in CID ms2 of m/z453. there must be m/z 340, m/z 566 and some other constituents. these comtaminatants come from your filter. please find the cid spectrum of m/z 453 in the attachment. 

in addition, these comounds should be decomposed polyamide.