So, after many, many man-hours with three different Bruker reps, as well as the consultant who set up the cold trap, we're ready to run. Nearly every part in the MS was replaced, finally the entire source module was replaced, and the MS came back to life; we speculate the source was completely coated by column bleed, possibly due to air leakage, though we were never able to find a leak. I had cleaned the lenses, but not the source body, so perhaps the body had active stuff on it the was "eating" all of our ions. In addition, we determined that another source of contamination was Viton ferrules in Bruker's plumbing outgassing some fluoropolymer. All of these were replaced with Valco furrules, and Valco "getters" were placed after all system pneumatics. This eliminated large, ugly, early-eluting peaks.
This uncovered a new problem, strange peaks of mostly m/z 52 and 56 were coming out around the same time, but were covered by the fluoropolymer peaks. After much experimenting, we solved this one, too. Nitrogen is our sample gas, and, somehow, when N2 from the cold trap goes on column, it is causing something to come off. Purge the trap with He before raising the temp, et voila, no more peaks.
So the question is, how could nitrogen possibly be reacting with the column? We're happy to have it working, but completely baffled as to why.