oscillation on splitter plate (CFT)

We were having many issues the last months with our GC-7200 qtof system. Many have been resolved, but we are left with tailing on later eluting peaks (phthalates, starting at about 250 °C) and oscillations coming from the splitter plate. 

Our system is as follows:

KAS-inlet (Gerstel), 15m column - backflush - 15m column - splitter plate - FID/qtof

The tailing is only on the MS side, and not on the FID. Oscillations are on both channels (and on the actual pressure reading of the corresponding EPC).

We extensively checked for leaks (with a duster), so splitter plate is good. I checked the inertness of the inlet side (including backflush) with the GC degradation mix from Restek (I know that it shows me leaks up to the backflush), so inlet/backflush is fine too. 

Also, analytical columns are in few months only and the system has hardly seen any samples. Transfer line has been exchanged again last week and tailing stayed. (Again, resolution is perfect on the FID!) MS source has been cleaned several times and hardly any samples have been on it since. 

Any ideas what the oscillations on the splitter plate are coming from and how to resolve it?

Any ideas on the tailing of later eluting compounds? (Could it possibly be the transfer line that's not heating evenly? How to check for this?)

Parents
  • Oscillations - like what?  Pressure reading?  Retention times?  Vacuum readings?   What units and how much?    Is it a purged splitter plate?  If so - - is the purge supplied by an Auxiliary EPC module with the required bleed restrictor mounted on top of the oven ?

    Tailing can be cold spots, too, especially for higher boiling point compounds. Is the GC pushed up tight to the MS transferline so that there is little to no gap ?   What is your transferline temperature?  Removable Ion Source temperature?

  • Vacuum reading of quad vac is stable at 3.26e-05 Torr, TOF Vac is not super stable, probably since the new turbo pumps (replace 2021 and 2022), 2.20e-07 to 2.4e-07. (not sure about the stability of the TOF vac, might have been the same before, but we don't check when everything is fine...)

  • Vacuum gauges and pressure readings are not always stable even when the vacuum or pressure is stable.  The vacuum gauge is reading very, very few ions of random gas molecules that bounce into the gauge, get ionized by electrons from a hot filament, and are then captured.  The TOF vacuum reading tends to be slightly less stable appearing than the Quad vacuum reading as there are so few molecules in there that the gauge has a more difficult time.

    Are the backflush Purged Ultimate Union and the Splitter plate supplied by EPC channels -- a channel off of a three channel Aux EPC module or the channel off of the collision cell EPC module -- or are they supplied by a PCM - a pressure control module?  If it is supplied by an EPC channel, there should be a Tee and a bleed restrictor mounted in place of that 1/16" union for both the PUU and the Splitter.   The EPC module must have enough flow going through it to control the pressure properly and bleeding off some is the only way.  A PCM supplied channel does not need that bleed restrictor.   Please contact Agilent about that.  

    I looked again. That coil of tubing top right is a bleed restrictor, so you have the right setup on at least that channel.  What is that at the red arrow? Make sure it's not capped off. It looks like a piece of tubing to make it easier to connect a flow meter. It must always be open.

    I just read from the top again. What are your column 1 and column 2 flows?  before and after the PUU backflush device?    Column 2 must be at least 0.2 or more ml/min higher than column 1.   How are you calculating the split ratio between the two devices?  How did you choose the lengths of the restrictors?

    The oscillations in your chromatograms above are not pressure, but signal. Please uncheck the FID signal so that only the MS TIC Scan signal is shown, then zoom in on the area of oscillation and share that chromatogram.  --- or to speed this up we need to figure out how I can look at your actual data file.

  • The oscillations are also in pressure. 

    Oscillations are on FID, TIC and Pressure Actual of EPC4. 

    I have removed the plastic cap from the bleed restrictor. (It has been on since initial installation, 8 years ago.) It's the backflush (EPC6). 

    Both backflush and splitter plate are supplied by EPC channels. (according to that the control panel calls the channels EPC4 and EPC6, they would be called differently if they were PCMs, no?)

    Column 1 is 1.2 mL/min and column 2 1.4 mL/min. I have calculated the split ratio with the provided Excel form: splitter3calc-rev1 (initially for another splitter plate installed on the instrument without makeup flow, we then added the splitter plate with makeup flow but kept the dimensions and it has worked 'til last autumn)

  • The scales are so small - 10^-2 and 10^-3. Are you sure that this is an issue and not that the signals are all so small that you're seeing it zoomed in?  I've never plotted EPC control down at that level to check.  The 7200 data files I have show wavy baselines at that scale.    Can you compare it with data files from earlier?

    Here are two FID signals with the only baseline sections I could get onscale at x10^-2...and you are showing FID at 10^-3.   This is using MassHunter Qualitative analysis Version 10.0. What software version are you using?

     

    All three signals show the oscillations which is a great clue.  What is the supply pressure of helium?  It has to be >20psi higher than any possible setpoint at the back of the instrument after any traps.  If the gas traps are older then the supply might need to be higher.  That's at least easy and quick to check. I set the helium to 80psi on the tank gauge as there's usually a mark there.

    After that I would guess that there's a tiny possibility that there's an EPC issue. Doubtful, but possible. We're getting close to where a visit by a Field Service Engineer might be faster.

Reply
  • The scales are so small - 10^-2 and 10^-3. Are you sure that this is an issue and not that the signals are all so small that you're seeing it zoomed in?  I've never plotted EPC control down at that level to check.  The 7200 data files I have show wavy baselines at that scale.    Can you compare it with data files from earlier?

    Here are two FID signals with the only baseline sections I could get onscale at x10^-2...and you are showing FID at 10^-3.   This is using MassHunter Qualitative analysis Version 10.0. What software version are you using?

     

    All three signals show the oscillations which is a great clue.  What is the supply pressure of helium?  It has to be >20psi higher than any possible setpoint at the back of the instrument after any traps.  If the gas traps are older then the supply might need to be higher.  That's at least easy and quick to check. I set the helium to 80psi on the tank gauge as there's usually a mark there.

    After that I would guess that there's a tiny possibility that there's an EPC issue. Doubtful, but possible. We're getting close to where a visit by a Field Service Engineer might be faster.

Children
  • We could resolve the oscillations! The gas was wrongly configured and setting it back, without changing anything other than that, did result in a much smoother pressure line and oscillations are no longer seen on the FID and qtof trace. (I don't know how this misconfiguration had happened, but I'm happy that the issue has been resolved.)

    pressure actual from EPC4 (purged 2way splitter), above: N2 set as gas, below: He set as gas (we use Helium)

    here are the FID and MS traces of these same samples. 

    The tailing problem on the MS is currently taken care of from our Swiss service technicians. 

  • The scale of your chromatograms is incredibly low, surprisingly low.  Most would never have seen these tiny perturbations.  Glad it's resolved.

Was this helpful?