Poor Consistency on Agilent 5975 GC-MS

[hallen]

No I've never used xylene/toluene to clean the instrument. It's prevalent in the large peak eluting around 21.8 minutes...

[01310040231]

If you have not been able to eliminate Xylene by now, I would clean/ replace the carrier line. How old is your split trap? Depending on the usage, you should change this trap every year.

[tlahren]

I would agree with 01310040231. I would also replace the copper line leading to the trap or remove it and clean it. I usually use MeOH, Acetone then Hexane and then Acetone again. After this then heat it gently to remove residual solvent. This has worked for me to remove almost "tar-like" contamination in the split-vent line and the system comes back to normal operating conditions. I'm fairly certain this is where your problem is.

I have a hard time believing that the peak is actually xylene/toluene as they would typically elute within the first 5 min on a 30m column. It is probably a high MW contaminant stuck in the split-vent with a phenyl group on it somewhere to be fragmented giving the 91 m/z tropylium ion (typical in phenyl-containing compounds). As you are doing a splitless injection you would be pseudo-injecting any contaminants from the split-vent line onto the column during the splitless time. When the splitless time is done you now have a large flow going through the splitvent and into the trap. This is why you still get a peak with a decent shape vs just having a large baseline of m/z 91 with no peak definition.

My rule of thumb for HP/Agilent GCs when I have reproducibilty issues or ghost peaks is to make sure the entire injection system (including the split-vent) are completely clean.