by
lmh » Mon Apr 27, 2009 2:29 pm
More segments shouldn't decrease sensitivity.
You have two issues to address, sensitivity and scan speed.
Are you using SRM, or running full scan MS2 events for each parent ion? If full scan, you can speed up enormously by choosing SRM, or manually choosing a very short MS2 scan around the region of a useful diagnostic fragment.
If your scan speeds are still too low, check you don't have a very long time specified as the maximum trap filling time. The instrument will wait this long even if there are no ions at all, so having 8 events waiting that long will really mess up your cycle time at low abundance!
If you are struggling with sensitivity:
(1) tee a standard solution into a typical flow rate of a typical solvent as you expect to use in chromatography, and use this (a) to optimise the spray chamber arrangements; (b) to tune for maximal signal.
(2) you can alter the automatic gain control target to get a bigger signal, but at expense of scan speed, and ultimately at expense of mass resolution. If you go beyond a certain point, the ions in the trap will begin to interact electrostatically and all resolution/trapping will be lost.
(3) check you don't have abundant contaminant/solvent-related ions present. These occupy trap-space and reduce sensitivity of target ions. There are various things you can do about them.
About arranging segments: remember that you do not have to put all 8 targets in separate segments. You can split them amongst any convenient number of segments depending on retention time, and you can even include one peak in two adjacent segments if it elutes inconveniently close to where you would like a segment boundary. For instance, if 5 peaks elute in rapid succession, you can have two segments, the first analysing the first three, the second handling the last three, with the middle peak in both.
Remember, also, that SIM will still give you a bigger signal than SRM in an ion trap. If your sensitivity problem is that you have run out of signal, SRM may give you phenomenal S:N ratios (because there is no noise!), but still give less reliable measurements (because there isn't much signal either!).
