GC-MS air percentage increases with oven temperature

[eoms]

We have an Agilent 6890N coupled with an Agilent 5973 MS. I recently installed a new column, and am trying to locate an air leak. I've checked all connections, made sure they are tight but not too tight.

I noticed that when I increase the oven temperature, the % air goes way up, but I don't understand how that could be related, except that it is on the GC side, not the MS side.

Any suggestions?

[zegockie]

On one occasion I had even more interesting experience: when I open an oven door I observe a huge increase of oxygen/air. Both nuts were tight.
I repeated this many times (oven temp was not high ~50C) - results were pretty much reproducible. Probably there was a crack somewhere, but I could not locate it.
Finally I fixed this by trimming both ends and re-installation of the column.

[Peter Apps]

We have an Agilent 6890N coupled with an Agilent 5973 MS. I recently installed a new column, and am trying to locate an air leak. I've checked all connections, made sure they are tight but not too tight.

I noticed that when I increase the oven temperature, the % air goes way up, but I don't understand how that could be related, except that it is on the GC side, not the MS side.

Any suggestions?

Check for actual leaks, not just whether the nuts feel as if they are properly tightened.

Peter

[rb6banjo]

I keep a can of that keyboard dust spray in my lab for this. You can go in and start the analyzer to scan or sit on the molecular ion for the fluorocarbon in the spray. Then, just spray it on your fittings. When the signal goes crazy, you know you've found your leak. It's easy and it's clean.

[eoms]

This morning, I did another air/water check and got values that were still between 5-10%, even after pumping over the weekend. I checked for leaks with a leak tester and compressed air, but did not find any.

I then just did a couple of blank solvent injections, did another air/water check, and the values have dropped to < 1.5%.

And while I typed this, before hitting 'submit', I checked one more time, and N2 is suddenly over 10%... Water is still around 1.5%

[kubowicz.tomasz]

Hello

Please check your carrier gas trap. I think it might be saturated and needs to be replaced.
Install new trap, give the GC column good flush and run autotune to check air/nitrogen/water.

Regards

Tomasz Kubowicz

[chemstation]

Hello,

Did you condition the column after installing and before connecting it to the MSD, and that you are now seeing the leak?
or
Is the MSD configured to have Chemical Ionisation, as sometimes Air& Water can get into the CI
plumbing, so you need to flush out it out.

kind regards
Alex

[eoms]

@Tomasz: our trap was replaced a few months ago, could it be bad already?

@chemstation: You are right, I did not condition the column before connecting it to the MSD, how could that be related? Is there anything I can do now? And we are using EI.

[Peter Apps]

Are you seeing oxygen as well as nitrogen on the air check ? If it is only nitrogen it could be that the m/z 28 is not nitrogen at all, but a contaminant from somewhere else in the system - there was some speculation on the forum that it might be column-related, which would fit with it being worse at higher over temperatures.

There have also been cases of contaminated helium - but then the n/x 28 should be constant.

Is the gas scrubber immediately upstream of the GC, and are there other instruments sharing the helium line ?

When you changed the column, did you do any inlet maintenance ?

Peter

[eoms]

We're seeing both N2 and O2. The gas scrubber is not sharing any other instruments and is about halfway the He cylinder and the GC. We did change the septum, liner and O-ring when we changed the column, and I replaced the septum and O-ring again yesterday.

[Peter Apps]

Try changing the inlet gold seal, and leak check the helium line and all connectors between the scrubber and the GC.

Peter

[aidnai]

@Tomasz: our trap was replaced a few months ago, could it be bad already?

Traps go bad based on how dirty the gas coming in is. If there is a leak upstream of the trap, that will shorten the trap life. (So check for leaks both upstream and downstream of the trap.)

If you find a big one and fix it, you'll also be saving yourself a bunch of helium

[eoms]

I've tested the line from the regulator at the tank to the instrument, and didn't find anything. I'll give it another shot later today.

@Peter: I can change the gold seal tomorrow.

[aidnai]

Test the regulator too, including cylinder connection and bonnet vent/leak ports. I have had regulators leak at NPT connections, at diaphragm seals, and of course cylinder connections.

[KipDM]

first: is your purge valve open or closed when this happens? if it is closed that may be the problem.

if you are using a septum make sure it is good and snug, if you are using a Merlin microseal septa replacement make sure to loosen the nut, slowly tighten while you watch your pressure gauge. when the pressure rises to the right level only turn 2 more "clicks" using the markings on your microseal cover.

loosen the front end nut (start of column) and then make sure the reducing nut is tight (and should have a metal o-ring in it).
make sure there is a graphite ferrule in the front end nut (or on the end of the column).

when tightening the front end nut make sure to NEVER loosen after done tightening.

check to make sure it is a hard (or non-deformable) ferrule on the rear end. we have had some rear end ferrules that were just a tiny bit too small and couldn't get a good seal.
when tightening the rear end tighten until you hear the squeal and then tighten two more 1/4-1/2 turns (more can be done but this gives good room for error).

i know this is basic too but also use a flow meter to determine that your flow rates and pressures are correct too.

if these don't work you can also cool down, break vacuum and check the gasket (if you had it open before the problem showed up) to make sure there is no lint or dust contaminant. if you see anything, or just to be safe, clean it and then put everything back together again.