Inconsistent BFB recovery on purge & trap GC-MS

[ruben dela cruz]

Hi to all, Perhaps someone can help me resolve my GC-MS dilemma. For several weeks now I have been trying to troubleshoot my problem on attaining a consistently passing surrogate (BFB) on a purge & trap, GC-MS (Tekmar, Varian3800-2000), but has been unsuccessful. I already changed my column, K-trap, filters, septa, and He source. Leak test is good, cleaned electrodes and housing, changed sample lines, baked many times, pass tuning. BFB recoveries on fortified blanks and samples always meet criteria, but when it comes to just blanks (boiled DI water), BFB recoveries are always very low and sometimes not detected. Also, I noticed that the very late eluting compounds have smaller peak response than usual although calibration is still linear. Has anybody encountered this kind of problem?

[James_Ball]

I have not run the instruments you listed, but I have run purge and trap. Many systems will have a problem where the more organic compounds present in the sample, the higher the response of the late eluting compounds. I don't normally see BFB affected that much, but Dichlorobenzene will be along with Trichlorobenzene.

It seems to stem from contamination on the moisture traps in the Encon Purge and Traps. If the Tekmar uses a similar cold tubing trap for moisture then it could be causing the problem. Over time higher boiling hydrocarbons can build up on the trap which will then retain the high boiling target analytes, with lower concentrations in the sample being analyzed being adsorbed more than samples with higher concentrations. If you can raise the temperature of the moisture trap and recoveries improve then that may be the problem and it just needs to be replaced.

[Bigbear]

Has your method ever worked well?
More information would be helpful. Column type and dimentions, flow rate, GC and P&T conditions.

I had a problem with BFB sort of like yours in the past which turned out to be all the water I was dumping into the source was activating it.
If you have not done so, clean your MS and then repeat your experiment.