Bruker 320 MS tuning in CI, accuracy problems

[Peter Apps]

I am running a Bruker 320 MS single quad in CI, with isobutane at 4 Torr.

Autotune (positive only) with the factory default FC43 tune compound gives masses within 0.1 Da of the target masses.

BUT, when I run a test mixture through the column the base peaks in the CI spectra are closer to the M+ than to the expected MH+, and even then are not accurate. For example ; indole 116.6 (MW 117), benzoic acid 121.7 (MW 122), 3-methylphenol 107.8 (MW 108), 4-methylphenol 107.5 (MW 108), 2-ethylphenol 121.7 (MW 122), phenol 93.4 (MW94), 1,4-butanediol 89.4 (MW 90), dimethylsulfone 93.4 (MW94).

This looks like a mass accuracy problem, possibly related to tuning.

Does anyone have experience of this, and is there a fix ? The manuals are uninformative.

Thanks

Peter

[JMB]

Peter,

1. All of your test cpds. are below true mass, therefore I think that you do have a mass axis calibration issue, as you believe.
2. FC-43 HAS MW 671, and i suspect that the calibration masses are mostly above the m/z 350 range under CI conditions.
3. Your test compounds have MW below 125 Da, so I think that the mass axis is lackig calibration masses in the range of tthe test compounds.
4. Can you do a manual tune or autotune under lens voltage settings that induce fragmentation of (M+H)+ to reasonably intense low mass ions ?

JMB

[Peter Apps]

Thanks JMB

I am not at the lab at the moment, but from memory the lowest mass in the CI calibration is above 200 Da, so I am sure that you are correct that the calibration is having to extrapolate to get to the mass range of my test compounds. The unknowns that I am really interested in getting a molecular weight for have MWs of about 130 - the potential molecular ion in the EI spectrum is very weak indeed.

I will certainly try a manual tune. The other approach that I thought might work is to tune in EI with no reagent gas - in the hope that the lens and quadrupole settings should be independent of the source of the ions.

Peter

[Peter Apps]

The problem was purely with the calibration.

The solution has been to set up a user defined tune compound in the autotune options, still using FC-43, but specifying three ions: 69, 131 and 219. Of these only 219 is on the default ion list.

Tuning the mass calibration to those three ions provides test compound base peaks that are exactly MH+.

Is there a similar problem with other instruments ?. In my Googling I noticed that one of the Agilent MSs tunes with methane and then switches to isobutane without changing the tune, which suggests that tuning with isobutane might be a general problem.

Peter

[JMB]

Peter,

I'm pleased to see that it was a simple fix; why couldn't the software be written to include/exclude low-mass ions at the click of a mouse ? But nice that you could build your own tune file.

Since CI (i-butane) does appear to give the m/z 69, 131, 219 ions I would think that the problem will occur across all instruments.

I had thought that maybe methane did, but butane did not, give these ions due to a difference in energy deposited in the FC-43 molecule (enthalpy ??)

JMB

[Peter Apps]

Peter,

why couldn't the software be written to include/exclude low-mass ions at the click of a mouse ?


JMB

The vagaries of instrument software are a source of continual wonder and not a small amount of frustration !

I should have thought that the most straightforward default would be to include all the ions - the instrument has no problem at all detecting the lower mass ions that I added. It could be that the lighter ions were coming from EI fragmentation rather than from CI, in which case there might be some conditions under which they would be too weak to be useful.

Peter