VOA chromatography and response issue

[cmbonner]

I have been experiencing an issue with the Dichlorodifluoromethane.

First, the peak is chromatographing unsymmetrical.

Second, if you break the peak down in to two sections and get the area response, A + B doesn't = C. When quantitating the unsymmetrical part of the peak i get a response of 8800 and when i quantitate the "normal looking" part of the peak I get 312000. But when I quant the two parts together the response is 510000.

I have changed the entire P&T (including transfer lines), column, inlet liners, adjusted split ratios from 20:1 to 40:1 and reinstalled software.

I believe this has to do with the CO2 peak, since that peak is chromatographing strangely also.

When I direct inject I don't have the problem with DCDFM but when the air comes out after the gaseous compounds, while much smaller than in a normal purged standard, it gives a unsymmetrical peak.

The issue is rather difficult to explain but i would be more than happy to email a file of the standard that i ran today if anyone would like to see.

Thanks in advance

Chris

[PeterB]

A couple of thoughts which may or may not be relevant.

Are you running an Agilent System in SIM?
Some Chemstation versions have a bug that misrepresents the area of a peak that includes the first or last scan of a SIM window. This is the only example I can think of where the areas of the two halves of a peak don't come even close to the area of the whole. The solution is just to not include the edge scan of the sim window in the area.

What is the starting temperature of your GC run?
Dichlorodifluoromethane is a gas at room temperature, so without cryo cooling, you aren't focussing the compound at all on your GC. Your peakshape is coming completely from the desorb of your P&T trap. How your compound is trapped (purge flow rate, trap temperature) as well as desorb conditions can all affect how the dichlorodifluoromethane comes of the trap. If you don't focus it afterwards, it's shape isn't going to get any better.

[Bigbear]

Depending on your sample volume and column size you may not be splitting enough.
I purge 25 ml. and split 70:1 using a 0.18 dia column.

[Yama001]

If the trap is not heating correctly, the gases can come off of the trap in a split peak as they move off the trap sections at different speeds. I see this more often with the trichlorofluoromethane - how do the other gases look?

[cmbonner]

Thanks for all your info

Peter - I am not running in SIM, this issue is in scan mode. I also don't have cryo cooling, using 5890s with 5972.

Bigbear - This occurs with a 5 ml sample volume running a 50 ppb standard.

Yama - all the other gaseous compounds look just fine.

A couple of notes.

I change the the purge temp from 57 to 55 and 62. This did change the peak some but i still get some variation of unsymmetrical chromatography.

I have three gc/ms systems that are identical (i.e. column, split flow, trap, etc...) and im not seeing this on the other two systems.

Thanks

[Bigbear]

Usually a fronting peak is due to column overload. Is that what you see? A peak that starts out looking fine on the upslope then is wider on the down slope?
Every gc/ms is a little different even if they are the same model configured the same. A quick test of the overloading theory is to run your 50 at a 60-70:1 ratio.
Purging at the higher temps should make your problem worse as you would be putting more water in your system. The gasses elute with the water causing band broadening.
Purge at ambient temp unless you are concerned with water miscable compounds.
What is the condition of your inlet? Liner and seal been in for a long time? Do you ever do direct injections? Septa particles in the liner?

[cmbonner]

For those that would like to know what caused this. It appears that the water source we have been using has a very high alkalinity, adding additional CO2. This was not caught at first because my other 3 machines didn't provide the same chromatography issues and it hasn't been happening the entire time we had been using the water. Apparently, the instrument has changed over time. Thanks for your input. Nothing like learning something new.

[Yama001]

That is fascinating. So the increased CO2 was suppressing the ionization of the freon?