Methamphetamine - poor peak shape

After a failed methamphetamine quantitation, it was noticed that the methamphetamine peak shape on our GC/FID instruments has been consistently poor or deteriorating through out a run.  This has been happening on both of our GC/FID instruments (all 3 columns).  I've been hard at work on the problem for over a week now and have tried a multitude of fixes, to no avail.  Any help/insight is appreciated.  Below is a synopsis of the things I've done as well as data.

 

Instrumentation:

7890B - 15meter HP-1   - labeled as GC1 on data

6890N - 15meter HP-1 - labeled as GC2 on data

 

In the past week, I have:

1.  replaced GC columns with brand new columns

2.  GC1 -

      changed the syringe

      changed merlin

      changed liner and o-ring (multiple times)

      changed gold seal x 2

3.  GC1 - swapped out autosampler towers (no change)

5.  GC2 - changed the liner and o-ring  (this instrument gets little use, so the gold seal is new)

 

Data

On GC1 - the GCQA06 standard methamphetamine peak shape looks fine on the printout, but when you zoom in it shows significant fronting.  This is the same QA std that we run routinely on our GC/MS instruments with no problems.  (see page 1-3 of attachment 1 (GC/MS instruments also have an FID))  All other peaks in the standard are symmetrical.

 

In attachment 1 see also data for our methamphetamine std, which is also used routinely on our GC/MS instruments with no problems.  Again, poor peak shape.  This standard was run multiple times on GC1 and that data is in attachment 2.  The peak shape continues to broaden to the point of splitting in Run 9, then actually "improves" to a shoulder for runs 10 and 11.  The peak area counts within this run also tend to vary.

 

On GC2 - the same methamphetamine standard was run multiple times.  As can be noted in the data from attachment 1 the y-axis abundance continues to decrease while the area count stays consistent between runs, indicating peak broadening.

 

*GC1 also houses a 2nd column, which is a 30meter HP-5 column, also new. The fronting meth peak in the QA standard is also seen in this system.  

* the methamphetamine in both standards is extracted

attachments.zip
Parents
  • The 11 injections are made with the same vial? I suggest to perform 11 run with 11 different vials, once the vials is perforated the solvent evaporate. I don't analyse drugs but 100:1 split lead to higher error...maybe you can inject less than 1uL and less split factor?

    The bad peak shape suggest phase activation, it'ìs better to add a precolumn made of simply silica glass.

Reply
  • The 11 injections are made with the same vial? I suggest to perform 11 run with 11 different vials, once the vials is perforated the solvent evaporate. I don't analyse drugs but 100:1 split lead to higher error...maybe you can inject less than 1uL and less split factor?

    The bad peak shape suggest phase activation, it'ìs better to add a precolumn made of simply silica glass.

Children
No Data
Was this helpful?