Poor sensitivity in CI GC/MS

[s.eaton@ucl.ac.uk]

I have a DSQ II with CI mode, which I use with methane. Recently, I have had a dramatic decrease in sensitivity in CI mode, especially negative ion, despite clean source parts (which behave well in EI) and a clean CI source. The user manual suggests using a needle to bend the filament wire close to the entrance hole. I am going to try this, but does anyone else have experience of poor sensitivity in NCI on a DSQ II?

Thanks,

Simon

[Peter Apps]

I have no experience with the instrument you have, but with a Varian/Bruker 320MS I got poor CI sensitivity that rapidly got worse - an electron multiplier close to the end of its life was a complicating factor. I would be really interested if shifting the filament made a positive difference.

Peter

[JMB]

I'm not familiar with your instrument, but the sensitivity trouble spots in GC-MS are generally, in order in which encountered by the analyte molecules,

1. dirty injection liner
2. dirty first inch of column (or not a clean cut across the tubing)
3. dirty last inch of column (ditto)
4. cold transfer line
5. sagging filament (or heavily oxidized)
6. low CI gas pressure
7. dirty source
8. dirty focussing lenses
9. dirty rods at entrance to quads.
10. dying EM

5) and 10) have already been mentioned. Did I forget the septum ??

[Peter Apps]

I'm not familiar with your instrument, but the sensitivity trouble spots in GC-MS are generally, in order in which encountered by the analyte molecules,

1. dirty injection liner
2. dirty first inch of column (or not a clean cut across the tubing)
3. dirty last inch of column (ditto)
4. cold transfer line
5. sagging filament (or heavily oxidized)
6. low CI gas pressure
7. dirty source
8. dirty focussing lenses
9. dirty rods at entrance to quads.
10. dying EM

5) and 10) have already been mentioned. Did I forget the septum ??

I compared EI and CI directly by switching only the source ion volume and turning on the the CI gas (iso-butane) between consecutive injections of the same test standard. I cannot recall the difference off hand, and I am not at the lab now, but it was enough to need extra concentration of the sample and larger injection volumes to see useful peaks in CI.

Peter

[s.eaton@ucl.ac.uk]

Thanks for the tips. I have done some more work on this with the following results, if anyone else experiences the same problems.

(i) filament. If the burn marks are centred on the hole in the CI ion volume when you remove a dirty ion volume, then the filament is positioned OK. This is more critical in CI because the hole is smaller. In my case, the filament was centred fine.

(ii) with the DSQ II, there is a little bit of 'play' in how the ion volume is positioned, which can be used to improve sensitivity. With CI gas on, and the CI ion volume at a stable 225 degrees, take off the front panel and top panel (so you can see where the tool is) put the calibration gas on and scan 60-650. Note the response for 452 and keep the calibration gas on. Insert the ion volume tool through the front of the machine, using the usual foreline evacuation procedure as for changing ion volume. Now using the tool, optimise 452 response by fractionally moving the ion volume position. This is quite tricky to do and can be irritating if you find a 'sweet spot' with the tool holding the ion volume in the locked position, then release the ion volume and lose the sweet spot. You can get a good increase in sensitivity by trial and error here. I was able to move the ion volume by having the tool part way between the locked and open position, find a sweet spot, and then release the volume. My 452 response increased nearly 10 fold by this procedure. However, some of this I suspect might be because the source has more play in it than usual as the two screws which hold the source in place are not completely screwed tight - I'll correct that next time I vent the machine.

(iii) the Thermo engineer suggested to use cotton buds with alumunium oxide paste for source cleaning as per usual, but to put the cotton bud into a Dremmel and clean at low speed.

Simon

[Peter Apps]

A ten times increase in response is impressive !. How far did you have to move the volume - I am assuming less then a mm ? The ion volume in the 320MS clips into place with two spring-loaded detents, so I am not sure that I will be able to use your technique, also it has a very simple lead through for the insertion tool that I doubt would be leak tight enough to run the MS while jiggling its insides.

Thanks Peter