I'm experiencing repeated problems of carry over when analysing PBDEs and dioxins using GC-MS (HP6890, Gerstel CIS4). I need to clean the injector (flush body with solvent, replace liner and sonicate septum less head in toluene) after every dozen or so samples to avoid the carry over. However recently the sensitivity for higher mass analytes has decreased significantly. Has anyoner else experienced this?
10ul injections (nonane keeper)
Gertstel CIS4, baffled deactivated liners. PTV program:
40c for 3min, 12C/sec to 300C hold for 4 miutes
Split 3minutes, splitless 3minutes, split to end of run.
By Anonymous on Sunday, May 9, 2004 - 09:14 am:
What is your speed of injection?
Do you inject in solvent vent mode with a MPS2?
If you inject with a fast autosampler (hp 7683) i think that your injection volume can cause the problem,try to reduce it.
Baffle deactivated liner cannot retain sufficiently your 10 uL,compare to glasswool or another packing.
What is your solvent?The contamination looks like backflash,although your temp is 40°C at start for 3 min to ventt a maximum of solvnet,perhaps when you start fast heating backflash occur,then 10 uL is big,and the liner capacity of a cis 4 is as small as 150/200 uL.
For the decraese in sensitivity for the high mass,it looks as if your lenses are dirty.
Try to wash it.
Are you gc under fume hood,for security.
Have you charcoal filter and....to prevent labor contamination????
By Anonymous on Wednesday, May 12, 2004 - 07:48 am:
You are trying to analyze some relatively heavy, low volatility compounds. I am not sure that 300C is a high enough temperature to vaporize these compounds completely. On some PTV inlets there is a significant lag time between the time the temperature sensor reaches setpoint and the time the liner temperature reaches the setpoint. I think that your starting temperature is probably ok, even on the low side if the liner is really at 40C. You don't give the split flow after the splitless time, the flow may not be high enough to flush out the heavy compounds, especially if the inlet is not hot enough.
I have injected up to 105 uL (using a different brand of GC and injector) without carryover issues, so 10 uL should not be an issue, although as pointed out above injection speed does matter.
Your source and lenses msy be contaminated or it may be primarily injection issues but I would go ahead and clean them to see if this makes a difference. The Agilent does not seem to tolerate a light deposit of material on the repeller as well as some other systems, so you will probably have to clean on a regular basis.
There have been report in the literature of PDBE decomposition at injection port temperature of 300C or less, but I have not seen any evidence of this myself. The response was lower for decabromo than I would have liked, but I did not find any peaks in the chromatogram that were reasonable decomposition products. Others I have talked to do not believe thermal decomposition at normal injector temperatures is major factor, but activity in the injection port may lead to loss of analyte. I would say the question is not completely resolved yet, but it would not hurt to try going to a higher injector temperature to see what effect this has.
My experience is that PTV inlets from various manufacturers all behave differently in terms of optimal temperatures and flows, so it is difficult to take a method from one system and run it on a system with different components without making changes to the method.
By M. Mandjikov on Thursday, July 1, 2004 - 11:21 am:
Which PBDEs are problematic? I do not see any carry over problems in my analysis but I am analyzing at very low levels -- < 250 ng/mL. I only use 2 uL injections.
By Anonymous on Thursday, July 1, 2004 - 01:07 pm:
The heavier (octa and above) are the more difficult to achieve complete vaporization and transfer to the column. These are the ones that also should be the most likely to exhibit carryover.
By ingochrist on Tuesday, July 6, 2004 - 06:51 pm:
If you are tired of the septumless head and want to get a septum head again, you can get one for 350 Euros at SIM GmbH in Europe or CHROMSYS for $450 in the US.
By ced on Wednesday, July 7, 2004 - 12:24 pm:
What are the problems with septumless head???
By ingochrist on Monday, July 12, 2004 - 01:12 pm:
If you handle the septumless head correctly and always maintain it properly, it will work just fine.
Make sure you do the following:
.. make sure there are no other gauge syringes than 23 in your lab
.. clean (sonicate) the head every once a while (while taking out the inner parts (o-ring swells!)
.. change the teflon needle guide once in a while
.. have some of the (really expensive) viton o-rings around. As far as I remember they cost more than $100 a piece.
The problems I have with the septumless head are:
- The head is so cold, that it freezes my analytes in the needle while injecting (headspace). That doesn't happen (at the same conditions) with a setum head.
Carryover is therefore a much bigger issue on large volume injection or cold-headspace injections or thermal desorption.
- The o-ring swells and can't be removed any more from the head -- continous sealing problems.
- the graphpack seals don't seal around the liner (that is great if I need high temps, but bad if I do splitless injections).
Let me know if you are ready for a switch
By elsa on Wednesday, July 14, 2004 - 03:07 am:
"...make sure there are no other gauge syringes than 23 in your lab " you could use a 23 or 26 gauge depending on the needle guide,keep in mind what you do...
"...change the teflon needle guide once in a while " as the septum with septum head...
"... have some of the (really expensive) viton o-rings around. As far as I remember they cost more than $100 a piece. " costs the two o-rings $80!!!
By Russ on Thursday, July 15, 2004 - 09:24 am:
We had problems with a septumless head when using a packed column injection port for megabore capillary columns. The valve did not seal the same after each injection allowing slightly more or less to leak past the valve after each injection. For the high port flows used in split injections this variability is not a problem. For lower flows, however, the retention time reproducibility was poor.
By ingochrist on Tuesday, July 20, 2004 - 09:52 am:
I do have a Gerstel PTV, and I like it, but I don't want to ever use the septumless head again. I still have to disagree with the usage of 26 gauge syringes; while they will likely work, you will scratch the o-ring and over time wear it out. The septumless inlet is designed to push down the plunger, which is above a spring. With a 26 gauge, you hit the edge of the plunger and have a 50% chance of either hitting the o-ring or the plunger.
Regarding the price -- I apologize -- the last time I saw that price was a few years ago.
By username on Monday, August 16, 2004 - 07:50 am:
The carry over issue has now been resolved. We replaced the PTV body; expensive but it worked and we have not seen carryover of the same like since. It may be that the PTV port metal body has become scratched or dirty through several years of constant use, thus generating 'active sites' increasing the possibility of carry over. Sensitivity of higher mass analytes has increased ten-fold since we replaced the body.
By GC user on Monday, August 16, 2004 - 08:06 am:
Thanks for updating us on the resolution of the problem.