Could be a number of issues. Have you tried injecting air? You'll get a nice Ar peak with center cup on mass 40-and it should give nice N2 too. If not, it's a leak or clog. Check open split and make sure capillary to mass spec is going where it needs to go and that it's not curling up on you. The tip of that capillary may clog or split-so examine it and maybe cut a wee bit off.
If you are set up with a "t" in the oven and a back flush valve, you may have a leak at the valve. The seal on the shaft in the valve, if the valve is in the oven, can open up-as can the capillary connections. These won't show up with Ar leak chasing, but water may show some bubbles. Another indicator is the bleed line-if the valve is closed, you should have He coming out the bleed line, but you won't see it if there is a leak there.
Nafion tubing may also pull loose or collapse/kink. Or maybe it needs to be cleaned. If the nafion is really dark on the inlet side, it's time.
Check the other bleed out lines too-quick check to see you're getting some He at least to that point.
Don't know of a convenient way to check He flow through open split. You can take them apart to check flow from the GC side, but that's when we tend to break them...Also, you might have a leak where the capillary enters the open split, and you may have to re-epoxy it if it's like our setup.
I take it you've done the standard Ar leak checking and checked all the usual suspects ;)
One other odd bit that can happen... If you are sticking a part of the capillary into the reactor from the GC side, a small piece might break off and work it's way down the tube. We've seen that on occassion doing gc/tc.
"Jennifer C. Stern" <[log in to unmask]> wrote:
>I've sorted through the archives and see a lot of troubleshooting of rising mass
>30 during EA-IRMS but I am having some issues with mass 30 spiking up at
>the end of a GC-IRMS nitrogen run and then slowly decreasing.
>The issue started with a routine reoxidizing of the oxidation reactor, after
>which there was the normal excess of 30 which was removed by solvent
>injections. But once mass 30 was gone, I still was unable to get nice mass 28
>peaks with my standard. They were all just blips. We have a quadrupole
>mass spec in parallel with our IRMS so I know that the compounds were
>coming out of the GC column just fine. So we replaced the reduction reactor.
>No real improvement, peaks barely discernible. There was a lot of water in
>the system so we thought maybe it was building up in the capillary and
>blocking the compounds (is this possible?) so we ran in carbon mode to make
>sure there was no blockage (carbon was fine). So over the weekend I made
>sure He flow was high to purge the system of water.
>Then we had a power outage (thunderstorm + building UPS failed) and I came
>in the next day and the turbos were off. So of course it took a few days to
>get the water out of the system. Now, I am seeing mass 28 peaks in the
>beginning of the run, but then I get a rise in mass 30 near the end of the run,
>and don't see any of the mass 28 peaks I should see. Running a blank of just
>my solvent didn't yield a mass 30 peak.
>Does this make any sense to anyone? Not sure what to do next...