HP-6890 FID Contamination

[Norm]

I am having a contamination issue with a 6890 FID. The contamination is light in molecular weight (c7-c15) and is low level (roughly 20x signal to noise). This poses a problem for my trace analysis where I am looking to 3x signal to noise.

I am running splitless on a 60m DB-5MS columns .25um, .25df. Hydrogen as a carrier gas and injecting 1ul methylene chloride. When I make an injection manual or automated the contamination is present. If I run a temperature program with no injection the contamination is not present. I have replaced everything from the gas source (now generator) to the inlets, EPC, liners, copper lines, o-rings, gold seals etc. with no luck.

It also appears that this is temperature sensitive. I am running the port at 280 however if I increase it to 290+ the contamination is worse.

Any help would be greatly appreciated.

Thank you in advance,

Norm

[Answer Man]

Norm - from everything you said, I would focus on septa, both the inj port septa and the vial septa. There are several ways to identify the problem:

1. Do you still see contamination if you inject 1µL of air? If you do, it is comming from the inj port septa or the syringe needle. If you do not...

2. Inject from a vial that you have not capped - do you still see the contamination? If you do, it may be from your vial and/or standard/solvent.

[Rick]

Do you use a liner for splitless injection or split injection?
1ul of Methylene chloride give 270ul of vapor...

maybe, you have flashback when you inject and the vapor of solvent clean the wall of injector.
try other solvent like ethyl acetate, use liner for splitless injection

Sorry for my english..i'm french canadian.

[xxx]

Do you use a liner for splitless injection or split injection?
1ul of Methylene chloride give 270ul of vapor...

maybe, you have flashback when you inject and the vapor of solvent clean the wall of injector.
try other solvent like ethyl acetate, use liner for splitless injection

Sorry for my english..i'm french canadian.

Salut moi aussi suis francais voila mon email
ledoeuff.cedric@rcc.ch

[semivoadude]

Do you have dual injection capabilities? If so, try connecting the column from Injector A to detector B, then evaluate.
Have you considered that the contamination may have come from the detector side? Maybe your jet is bent or your collector assembly is dirty
Could they be ghost peaks? inject a blank vial w/ nothing in it.
Have you tried a different column?


Good luck

[scbowman]

Norm -

Don't forget to test the solvent itself for contamination. We recently had a contamination issue that turned out to be, after changing and testing everything on the instrument, related to the hexanes we were using. Apparently, there was a brominated intermediate that was left over from the solvent manufacturing and present in a number of different vendor's hexanes. Try a different lot or a higher purity of MeCl2.