Persistent problem of high N2 in GCMS

[wan]

We have an Agilent GCMS system.

From time to time, after running some long sequences, we realised that the N2 and O2 values are high after we did tuning.

In order to bring these values down, we have to tighten the column nuts. The strange part is we did not change the column in between the runs/sequences. So how can the nuts loosen by themselves when we did not change anything, but were merely running analysis?

I don't quite understand how something that is in the oven can loosen without any manipulation/contact.

Is there anything I can do to prevent this problem from recurring?

[littlepeaks]

I suspect that the column nuts expand and contract somewhat with the oven temperature. Also, not sure if the ferrule shrinks a little bit with use.

We have had problems with Agilent column nuts sealing well in the last year. We found another company that makes column nuts that work a lot better. I probably shouldn't mention the brand name because of sponsorship for the forum, but there is at least one better column nut out there.

[Peter Apps]

This is a known problem with Vespel-based ferrules, and the installation instructions for the columns and instruments that I use (Restak and Varian) mention that fittings need to be snugged up periodically.

Do not be tempted to solve the problem by overtightening the fitting.

Petr

[bhuvfe]

Wan,

Some years ago I had the same problem, especially with fast ramping of the oven.
I had luck with metal-based ferrules such as http://www.sge.com/products/gc--lc-supp ... -ferrules3

The only drawback is that once they are installed you cannot touch them anymore (e.g overtighten them will not work) but once installed correctly they last for months.

Good luck.
bhuvfe

[Forse]

Normal ferrules work themselves loose when the GC oven ramps the temp up and down.

I have alsor used SGE metal ferrules in an Agilent GCMS system. Works fine.

[wan]

Thank you all!

I am perplexed because we use the same nuts/ferrules for another Agilent GCMS we have and do not encounter such problems.

I'll try the SGE metal ferrules and see if it will require less frequent tightening.

[bhuvfe]

Thank you all!

I am perplexed because we use the same nuts/ferrules for another Agilent GCMS we have and do not encounter such problems.

I'll try the SGE metal ferrules and see if it will require less frequent tightening.

Just for my curiosity.
Are you running the same method on the other GC-MS instruments?

[wan]

Yes I am, when I get to use both at the same time.

The only difference between those two GCMS is one has a CTC while the other has a ALS injector.

[Don_Hilton]

You may discover it is just the particular ferrule, transfer line end, or nut. I would suggest that you make sure to keep it tight. Do not try to reuse ferrules when you change columns (this is a general rule - not just for this instance) and see how things work with the next ferrule.

[labstephan]

Hello Wan,

we had a similar issue. I assume you are working with a 6890 GC and the CTC is not original incorporated into the Chemstation. Your GC-MS method uses a splitless injection.

Our CTC is controlled via a macro from Chromtech which uses simple start/stop and GC idle signals. So the GC is anytime in pre-run conditions without any split flow on. The SSL inlet from Agilent 6890 is known for "sucking" air from the gold seal in splitless mode ( so it was told to me!).

We solved the problem with a standby method, which is allways loaded, with a split flow of 50 mL/min.
Starting a sequence loads the needed measuring method and after finishing the sequence the standby method is reloaded.
The original Agilent ALS doesn`t show this problems. The pre-run conditions are only active just before the injection.

By the way the SGE Siltite`s are very nice ferrules, but don`t overtighten them!

Hope this information are helpful.
Best regards

Stephan

[wan]

Hi Don, Thanks. We do not reuse ferrules too. We are still monitoring the situation and will be trying out the SGE metal ferrules when they arrive.

Hi Stephan, the GCMS that is giving us problems is not the system installed with the CTC, but the system installed with the ALS injector and it is an original Agilent ALS injector. :/

[Don_Hilton]

Do take note of Stephan's point about doing your leak checking with enough split flow to keep air from diffusing into the gas stream through the split vent. This is one place where your identical instruments may not be quite identical.

[wan]

Hi Don, Thanks! I will take note of that.