Sudden flat peaks

Hello,

I have been running a method for residual solvents in the Headspace GCMS for a couple of months. Last week I ran a calibration curve, and this week i re-ran another one ( same standards levels and internal standard). But this last calibration curve i happened to get some peaks with flat tops, and the absolute response is almost double compared to other calibration curves I have run in the past. How is this possible? I have not changed anything in the method other than the SIM window due to change of capillary during the PM. 

After the PM I have run samples and calibration curves without this issue, this was a sudden problem and I cannot find the reason this could happen.

I have attached some pictures so you can see the issue. Any ideas would be wonderful! Thank you!!

   

  • The maximum abundance per ion on SQ instruments is approximately 8.4 × 106, To stay within the linear range of the electron multiplier it is recommended to be below 2.0 × 106 for SQ.

    Those peaks are saturating the MS.

    What is the EM setting in your method?  What is the EM in your last autotune?  Did something change in the GC method?

  • Hi Paul. Thank you for replying.

    In my method the EM Setting is Gain factor and the gain factor is 10. The Agilent engineer from the PM recommended to run a etune. I Have attached the results.

    Looking back at the previous method the applied EM voltage is 1311. On the new method the applied EM voltage is 1360. I don't remember changing that setting, and I am the only one that handles this instrument. But could this cause this issue?

    Thank you!

      

  • You said there was a PM so everything got cleaned? If so when I clean my MSD sensitivity goes way up and I have to lower the EM in the method and/or tune file the method references.

  • Gain 10.  Using Gain means that the EM will be adjusted after every autotune when the system calculates a new Gain Curve.  Gain 10 is quite high unless your application requires you to really drive the system. Few headspace methods need super sensitivity.  The lowest concentration of the smallest peak in your calibration standards only needs to be 3x to 5x the average baseline.  Your highest concentration of your largest peak in your calibration standards should be only about 1 million in SCAN mode.

    Turn the Gain down to 1 and run your highest standard. The first residual solvents application note I read 5991-2079EN.pdf   runs at Gain 1.   Or, look at your previous data files. Under the data file *.D subdirectory is a file called "acqmeth.txt."  Look in there and towards the bottom of the file:

    34.593 : EMISSION
    70.007 : ENERGY
    22.349 : REPELLER
    90.331 : IONFOCUS
    27.717 : ENTRANCE_LENS
    1124.702 : EMVOLTS

    1195.6 : Actual EMV
    1.00 : GAIN FACTOR

    The Actual EMV is what was used during that run.  To check it out you could change your current method to Absolute EMV and use the setpoint from the known good data file.

    The tune is beautiful.  Were you running ATUNE before and not ETUNE and changed your method to use that tune protocol?  ETUNE is more sensitive for masses higher than ~131 amu than ATUNE. That additional sensitivity could be making your peaks appear to be bigger than they were before as well.

    Then there are the possible GC issues.  Headspace is run using SPLIT mode. If the liner gets clogged, the copper split vent line gets clogged, or the Split vent trap gets clogged, the split flow may not be the same as the GC thinks. This could increase the amount of sampler on column.   The GC PM includes replacing the split vent trap and liner and verifying flow through the split vent line, so that is probably not a problem.

    and let us know.

  • Thank you so much Paul! I'll try to lower the gain and keep the etune. 

    I was running atune before the PM, after the PM its when we switched to etune. I ran some samples after the pm and did not have this issue, until this week. 

    Yes, the spilt vent was replaced so I don't think that could be a problem. 

  • There was a PM. After the PM i had to get new SIM windows since some analytes shifted and i run other test. During these test i did not see this problem. As Paul said it might be the switch of the tune file from atune to etune after the pm. However i did run some samples after this switch and i had no issues until this week. 

Was this helpful?