no reason for peaks

[scbowman]

Our recent dilemma involves a set of 4 peaks that started out as an irritation but have grown into a downright nuisance. We have done everything including called agilent to track down the source of the peaks. We currently run analysis on an HP 6890, but when trying a different instrument we saw the same issue which indicated it had to be either gas or solvent (the only thing common to both - and an air injection showed no peaks, i believe, excusing the detector...oh yeah, shows up with both FID and ECD). (Believe me we (my group and agilent) tried all the obvious tests, different lots of solvents, new columns, new syringes, etc.)

Complete replacement of the inlet system, gas tank, gas lines, traps, pressure modulators, etc. did not solve the problem. This led us to believe it was the solvent (hexane in this case). I should also mention that the peaks do not show up in acetone or methanol. We concentrated down the solvent and sure enough saw a concurrent concentration of the peaks. However, we had already tested 4 or 5 different lots of hexane and after that we tried a completely different vendor. Still saw the peaks. The peaks stayed constant for months with a small signal around 300Hz (on ECD). Then they disappeared for about 3 runs. Then they came back and have been increasing in signal ever since (currently 3000 Hz). Has anybody ever seen this apparently unresolvable issue before? Any ideas, thoughts, comments? Anything would be appreciated.

[gcguy]

Could it be plasticisers from your autosampler vial lids or septa? Have you tried other non-polar solvents? Do you have access to ms?

GCguy

[scbowman]

well, we see the peaks even with a manual injection and we have tried different inlet septa. even soaked them in hexane for a while to see if that increased the signal. it didn't. i believe that the service reps tried isooctane, but i can't recall off the top of my head if they saw the peaks or not. i want to say that they didn't because they all seem very focused on the hexane. the last i heard from agilent was that they would try the hexane on a GC/MS to see if they could identify what the peaks are. the only thing that bothers me about the hexane is that we saw it in 4 or 5 lots of hexane from Fisher as well as from a burdick & jackson hexane from a completely different lab. so, it seems to me that due to all the different sources of hexane that have been tried it has to be something soluble in hexane that gets carried on to the column. but with the completely new inlet system i can't figure out what that would be.

[Consumer Products Guy]

Have you replaced or really cleaned the lines and body of the inlet weldment? Have you physically removed the gold seal/washer and scrubbed out the inlet housing with solvent? I think your injection solvent is reaching some trapped contaminant somewhere in the inlet.

[scbowman]

the inlet housing, tubing to and from the inlet (all the way from the tank), gold seals, washers, liners, etc. have all been replaced. it is all new.

[scbowman]

For anyone who may have been following this topic: We have tracked down the source of the peaks to the hexane used. We sent a concentrated sample of our solvent off for MS analysis and it turns out that the peaks we were seeing were actually brominated species. The hexane we were using was for residue and pesticide analysis by GC. Another source of hexane (also pesticide) grade from a different supplier, and from a different laboratory, also showed the same peaks. It is believed that halogenated species may be used as intermediates in the production of hexanes. Therefore, it appears that resolving this issue simply requires finding the right source of hexane that contains low enough amounts of contaminants that they won't be observed using a u-ECD.