Bad tailing shape for early peaks only

Please include sufficient information.   What chemistry/peaks/sample/matrix?  What column with dimensions? What are all of the GC method parameters? 

These peaks are gigantic - millions of counts tall. Dilute 75% or more, if split injection double or triple the split ratio, if splitless - change to split.  Huge peaks contaminate the source faster and increase the need and shorten the timing for inlet and column maintenance.

Hi Paul, 

I injected verification solution of caffeine to compare and I had 2.2 of tailing instead of 0.9. 

Yes the peak are very high also, but I inject the same concentration. So cleary here I have a problem. 

I also found this Agilent Notes that tend to confirm the need of trimming at the end of column which is something I didn't perform yet. 

GCTroubleshooting_part_four_tailing_peaks.doc.pdf

I think I have column activity, because I saw tailing only on polar compound. 

So I will vent and trim at least 20 cm for the MS port. 

I will keep you in touch. 

Hi Paul, I think we had column activity at the MS connection. My tailing of my Caffeine peak was 2.2 before I trimmed the column at the detector and now it's 1.0. If we compared from our old data, usually is around 0.9-1.1 of tailing. 

But when I tried to reinject my solution show in this chat, I still have bad shape for early peaks. 

I ran a pressure decay test and it failed. 

Everything was tighten adequately, but pressure decay test failed again. 

I assume that this leak will affect more early peaks than lated eluting peaks. 

That might be the reason. 

The leak must be fixed.  Cool off the inlet and oven, cut the column, and then carefully press the cut end into the side of a septum to cap it off.  Then cap off the bottom of the inlet and run the decay test again. If it fails, replace the septum, liner and o ring, and gold seal, then cap it off and run the pressure decay test again. The inlet must pass this test before reinstalling the column.

You can cap the bottom of the inlet a few ways:  A no-hole ferrule.  A piece of column installed normally with the end pressed into a septum.  A small paper clip bent open can be installed with a ferrule just like a column. 

I check at the split vent cartridge and I turned half a turn clockwise. Nobody in my laboratory changed this cartridge so I don't why it was loose. The pressure decay passed after that. 

My early peaks are now ok. 

Is it possible with time that this split vent cartridge get loose and need to be retightened ? I also see that on another GCMS instrument in our laboratory. 

I also checked the flow at the split and it's 19.7-19.8 mL/min for a 20 mL/min. 

Before I touched the split vent cartridge  it was 18.6-18.8 mL/min  but since Agilent tech support in Canada it's around 10% I was thinking it was ok because 10% of 20 is plus or minus 2. 

So far the system looks fine. 

Thank you for your help.  

The split vent cartridge should be replaced about once per year or so. It is included in a preventive maintenance.  They don't typically loosen unless the temperature in the lab fluctuates a lot.

It will be extremely reproducible. If your flow meter is not calibrated yearly, it may not compare anyway.