Appropriate Gain Settings (SIM and MRM)

Hi All,

I'm in the process of optimising a new method for a trace level mixture using an Agilent 7000D. Whilst I am fairly confident with many aspects of the instrument, MS gain (and EM saver) are features that I don't have a lot of knowledge about and would love to know more as been using a gain level of 1 for all the work I have done so far. 

Whilst there is some literature online covering how to set the appropriate gain levels, this is a good few years old (referencing Chemstation) and treats gain as an attribute that is applied to an entire run, rather than something that can be applied to individual SIM or MRM time segments. With that in mind I would be hugely grateful for any response on the questions below if somebody would be willing to oblige?

Within the context of a SIM or MRM run, is it acceptable to use different gain levels for different analytes/SIM groups? Whilst I know that increasing gain is going to increase noise, my gain increase experiments so far do show an increase in SNR and I suspect that increasing gain will help to address some of the problems I am having with getting good integration with low levels peaks. 

How are gain levels set? From what I understand, we're looking for about 3 x 10^6 counts for SIM (and 10^7 for MRM) from the most intense ion... Is that correct?

Lastly, is setting the EM saver to 1X10^8 default acceptable and reasonable?  

Many Thanks


  • To have the least amount of ongoing instrument maintenance you want to inject the smallest amount of volume of the lowest concentration you can and still be able to see the lowest level calibration peaks at only 3x or 4x the amplitude of the background. So if the baseline is at 1000 and the baseline amplitude is from 900 to 1100, 200 counts, then your lowest cal peaks only need to be 600 to 800 counts tall above the baseline.  Too many inject too much, too high, wanting giant peaks in the millions and then complain when they have to clean the inlet, cut or replace the column, clean the source, replace source insulators, and do even more remedial maintenance of the mass spectrometer over time.  Everything that comes out of the column goes into the mass spectrometer. Some amount of everything that goes into the mass spectrometer gets everywhere and continuously redistributes, so starts to contaminate a clean source as soon as it is installed.  


    Have you read "5991-2105EN_5989_5672.qxd (" ?   Yes, it's nine years old, but the concepts are the same. I would guess that's the literature you are referencing.

    And this" "Optimizing Conditions for GC/MS Analyses ("    Slide 10

    Are you creating separate time segments or using dMRM ?

    Using different gain settings per time segment is not unusual.  Don't be surprised at the baseline differences with changes in SIM/MRM ions and also Gain settings.

    There is a dance with gain/EM and SNR/peak height and SIM/MRM or Scan.  

    - A high gain setting does influence the degree of non-linearity.

    - Compounds with higher absolute ion intensity are more likely to show non-linearity.

    In the case of MS/MS trace applications, many users have increased the Gain to amplify the response of less sensitive peaks at the lower end of the calibration curve. Although this may be appropriate at the LOD/LOQ for all compounds, the compounds with higher response will be more likely to demonstrate the non-linear response at the high end of the calibration curve.

    Whatever you do, run your high standards, and at least some samples once in a while ,in MS2 SCAN mode, Gain 0.5.  Evaluate the data looking for huge peaks or contaminants.    In a QQQ, ALL of the possible positive ions that can be created in the source and focused into the quad go into the quad.  In MRM mode, only the MRM Parent ion is allowed to go through – all the rest of the ions either crash into the quad or are pumped away.  This is the point of MRM, but also can cause problems if operators only think about their GCMS from the MRM viewpoint.  If they keep pushing the amount injected and increase the total sample concentration to see smaller and smaller peaks of interest , they will cause problems.  

    EM Saver:

    EM Saver is used to prevent a non-linear detector response at the upper range of a compound's calibration curve. This non-linear response can be caused by setting the gain value or dwell time too high on compounds with high ion concentrations. With EM saver, high gain values and dwell times can be used without saturating the EM.

     EM Saver is used with SIM mode in a single quadrupole configuration and MRM mode in a triple quadrupole configuration. A benefit of using EM saver is a more consistent detector response and an extended EM life.

    • EM Saver is available for SIM or MRM modes, those with dwell times – not SCAN modes.
    • During the ion or transition dwell time, the system accumulates ion current – it adds up all the counts that happen in that time period. If the EM receives excessively high signal during this time, the signal will be saturated.
    • EM Saver sets a maximum sum limit for every ion or transition.
    • EM Saver limits the accumulated ion current to the user-selected summation limit. Its use is designed to spare the EM from ongoing exposure to excessively high signal abundances.
    • EM Saver can help with GAIN values that are too large or dwell times that are too long, both of which will increase the number of counts over the selected dwell time.

  • Hi Paul,

    Thanks for a truly outstanding response. I very much appreciate the time you have put into your answer and am much obliged for you going the extra mile (and then some)!

    I have to admit that I have been slightly guilty of injecting a little too much sample myself in the past, but I have taken a bit more care recently, something our engineer apparently noticed in the most recent PM...

    Now that my work has gone from been scan based to SIM/MRM (and involves comparatively few analytes) I only ever use time segment methods and have yet to give dMRM a go...

    Anyhow, thanks for the brilliant response and have a great week.


Was this helpful?