GC/MS: high baseline, lots of noise

[gurlwithgcms]

Hi! I am having some problems with what seems to be an inconsistent high noise level the GC/MS I am working with. It is an Agilent 6890/5973 used for VOC analysis (EPA 524.2). I ran a BFB injection Monday and it had pretty nice peaks. Wednesday and today I ran it again and I get a lot of noise with a very high baseline. I replaced the column last Friday because I was not seeing any peaks using the old column. Between Monday’s and Wednesday’s BFB injections, I ran and air/water check, tightened column nuts, but did not do any other maintenance. I am fairly new to the GC/MS and would appreciate any advice about how to fix this issue, and what is causing this.
The good example (https://www.flickr.com/photos/129144858 ... 107489664/)
and the bad example (https://www.flickr.com/photos/129144858 ... 107490714/)
are uploaded on Flickr.
Thanks

[kubowicz.tomasz]

Hello

Abundance for "bad" run is very high. I'd check:
1.Gain factor for MSD (perhaps to high)
2.Threshold (too low)
Of course I assume that there is no water and air in the system

Regards

Tomasz Kubowicz

[dblux_]

Hi! I am having some problems with what seems to be an inconsistent high noise level the GC/MS I am working with. It is an Agilent 6890/5973 used for VOC analysis (EPA 524.2). I ran a BFB injection Monday and it had pretty nice peaks. Wednesday and today I ran it again and I get a lot of noise with a very high baseline. I replaced the column last Friday because I was not seeing any peaks using the old column. Between Monday’s and Wednesday’s BFB injections, I ran and air/water check, tightened column nuts, but did not do any other maintenance. I am fairly new to the GC/MS and would appreciate any advice about how to fix this issue, and what is causing this.
The good example (https://www.flickr.com/photos/129144858 ... 107489664/)
and the bad example (https://www.flickr.com/photos/129144858 ... 107490714/)
are uploaded on Flickr.
Thanks

Why not start with common PFTBA tune evaluation ?

[gurlwithgcms]

Thanks for your responses Tomasz and dblux_

I did another air/water check this morning, and it changed from yesterday. Yesterday it had 3.12% water, 4.95% nitrogen, and less than 1% for oxygen and carbon dioxide. Well, today it has less water (less than 1%) and more nitrogen (15.65%). I used a leak detector around the column nuts, inlet, transfer line, and diffusion pump, and pretty much everywhere else too, and there was no leak detected. The most recent tune evaluation (https://www.flickr.com/photos/129144858 ... 547510910/) doesn't look as good as it has in the past. The 502 peak has some stuff around the baseline, even though I cleaned the source last week, and I'm not sure why the peak looks split.

[Steve Reimer]

At 15+% nitrogen you have a leak. To look for it arond the MS use a can of Dust Off or other canned "air" that contains a fluorinated hydrocarbon. Depending on which hydrocarbon you have you will get a peak at 69m/z or something usable. Go to manual tune, parameters, and scan the mass spec looking for the appropriate ions in your canned air. To make it more obvious turn off the pftba and watch for abundances in the scan. Now spray the canned air around the side door and any part that may have been disturbed during your last service. Remember that it will take >5 minutes for anything done at the inlet to get to the MS.
When the abundance goes from a few counts to 1000000, you found your leak. I often find something in the door o-ring, cat hair or lint, anything like that can cause a leak.

Good luck!

[gurlwithgcms]

Thanks Steve, I'm in the process of trying to find the leak as per your suggestion... fingers crossed I find one

[dblux_]

Thanks Steve, I'm in the process of trying to find the leak as per your suggestion... fingers crossed I find one

Your last tune report shows 4.44% N2 and 0,35% O2.
Relatively to oxygen you have higher nitrogen content than in air. Do you use nitrogen as purge gas in P&T concentrator ? If yes, then 6-port valve may have leak to carrier gas port. Just to consider another possibility

[Peter Apps]

Have you changed the helium cyclinder recently ? - there are sporadic reports of contaminated helium giving the symptoms that you describe.

Peter

[kubowicz.tomasz]

Hello

I would check ion source...Check your autotune report - EntLens is 0.
You also have wrong isotopes ratio, high EMV.
To isolate GC you can plugged MS transfer line with blank nut and then perform autotune, it will help you to eliminate GC.

Regards

Tomasz Kubowicz

[PaulB]

Quick warning - most aerosol dusters now just contane butane (ozone layer friendly). I even had to point that out to an Agilent engineer who was happily pointing one at lots of hot fittings

[James_Ball]

Hello

I would check ion source...Check your autotune report - EntLens is 0.
You also have wrong isotopes ratio, high EMV.
To isolate GC you can plugged MS transfer line with blank nut and then perform autotune, it will help you to eliminate GC.

Regards

Tomasz Kubowicz

When running BFB tunes my EntLens is almost always 0, if it goes above about 2v then something is wrong. The Variable EntLens Offset makes up for the EntLens being 0.

[James_Ball]

Thanks for your responses Tomasz and dblux_

I did another air/water check this morning, and it changed from yesterday. Yesterday it had 3.12% water, 4.95% nitrogen, and less than 1% for oxygen and carbon dioxide. Well, today it has less water (less than 1%) and more nitrogen (15.65%). I used a leak detector around the column nuts, inlet, transfer line, and diffusion pump, and pretty much everywhere else too, and there was no leak detected. The most recent tune evaluation (https://www.flickr.com/photos/129144858 ... 547510910/) doesn't look as good as it has in the past. The 502 peak has some stuff around the baseline, even though I cleaned the source last week, and I'm not sure why the peak looks split.

How old is your Electron Multiplier?

The counts for 69 at 1.4mil are good for a BFB Tune but the EMVolts at 2635 is way too high. A leak can cause that to run high because the loss of vacuum will lower the sensitivity, or if you just have an old multiplier it is pushing the volts up to compensate and that can also cause funny things to happen with air and water not calculating corrrectly and sensitivity across the mass range which could make the 502 look low.

Another strange thing in that tune report is that in Profile it shows mass 69 at almost 3x the counts of mass 219, but the full scan specta shows them with almost the same counts while 502 is showing the same counts in either mode.

[Steve Reimer]

For BFB tune I have 800,000 for the target abundance for 69.
I also see that your repeller maximum is at 20 volts, it can be set to 40v. 20 volt limit is a leftover from the 5970.
Also the repeller and ion focus are both at the maximum values, not a good sign. With the EM voltage at 2600 it looks like your system can't find a good tune.

[RogerBardsley]

We have had issues with tanks containg carbon dioxide and high levels of argon. The last argon issue buried our peaks with noise, but when the specific ions were checked, the peaks were OK.