High initial baseline in GCMS

[donnygg]

I've been having problem with high initial baseline in my GCMS system which is reducing the detection limit in the range. I set the solvent cutoff at 2.5 min and it starts with a TIC of about 4x10^6 but takes 15min before it reaches 50000, which was the normal baseline I usually get. I started getting this problem after we shut down the system for renovation in the lab. I checked the MS and the main contaminant is oxygen (m/z 32) and nitrogen (m/z 28). However, the ratio is 1:2 (O2:N2), which the Agilent handbook says is a rough indication of a "leak-free" system. I thought it was leakage in the inlet septum but changing it did not help. The helium carrier gas is ultra pure and should not be a problem since the high background is not present throughout the whole run. Anyone has any idea what could be the problem? Thanks!

[Peter Apps]

This sounds very much as if your split line is clogged - can you verify that the spectrum during the period of elevated baseline matches the solvent that you are using ?

Peter

[donnygg]

The MS shows very low relative abundance of the fragments for my solvent (chloroform, m/z 83) because it is almost dominated by oxygen. The MS looks as if it took 15 min to clear out most of the oxygen. The abundance of m/z 32:83 is 2x10^6 : 2x10^4 at the start of the spectrum. Also, we are using the splitless mode.

Cheers
Don

[Peter Apps]

Hi Don

I presume that you open the split at some time after the injection is made - how long is this splitless period ?

What happens if you do a blank injection (zero volume in the syringe).

What are you injecting besides chloroform, what syringe are you using (volume and type), are you using an autoinjector or doing it manually ?

Peter

[donnygg]

I think it's splitless for 1 min. The "Purge to split vent" is set at 50mL/min at 1min.

I've not done a blank injection but hexane and chloroform both give the same high initial baseline.

I've been injecting alkane samples in chloroform while others are analysing derivatised fatty acid biological samples. We're using the Blue line autosampler syringe from Agilent and we use an autosampler.

[Peter Apps]

I cannot htink of any way that you could introduce enough air during an ordinary injection from an autosampler to give the effect that you describe, and if the air was coming from anywhere other than the injection it would be more or less constant.

It is possible that the 32 and 28 are from something other than air - what other ions are there in the tail ?.

What was changed during the lab renovations - did anything happen to the gas lines to the GC ?

What inlet maintenance have you done recently ?

Peter

[donnygg]

It definitely seems like air. There's m/z 28, 32 and 40 (N2, O2 and Ar). The rest are so small relative to these peaks.

I'm not sure if anything happened to the gas line. The area was wrapped under plastic sheets, so I doubt the workers touched them. Prior to the shutdown, everything was running fine. After restarting the system, ie. pumping down vacuum of MS, turning on GC etc., we did not conduct any maintenance. Then this happened. Maybe we'll just call in the engineers to have a look.

[Peter Apps]

we did not conduct any maintenance. Maybe we'll just call in the engineers to have a look.

New Years Resolution: Maintenance is necessary, maintenance is good, maintenance is a lot cheaper than service calls. Try changing the inlet liner and gold seal for a start - and if the last change was so long ago that nobody can remember when it was done, you might need to clean out the inlet body also.

Peter

[willnatalie]

Does this baseline increase with Temperature or is it completely independent? Your not seeing any water or CO2 at 44?

[donnygg]

we did not conduct any maintenance. Maybe we'll just call in the engineers to have a look.

New Years Resolution: Maintenance is necessary, maintenance is good, maintenance is a lot cheaper than service calls. Try changing the inlet liner and gold seal for a start - and if the last change was so long ago that nobody can remember when it was done, you might need to clean out the inlet body also.

Peter

The inlet liner was changed only last month. The rest, probably over a year ago. We're still under warranty, so calling for service should be free though it will take a few days to schedule.

Does this baseline increase with Temperature or is it completely independent? Your not seeing any water or CO2 at 44?

It decreased with temperature and time, since we use gradient temperature. It actually drops faster with steeper gradient. I don't see water (m/z 18) as it's out of my scanning range. I think there's 44 but it's much lower than 28, 32 and 40.

[Peter Apps]

Derivatized fatty acids from biological matrices can put a lot of muck into an inlet liner. Surely it is worth changing the liner and gold seal as a troubleshooting excercise ? I doubt that this kind of routine maintenance is covered by the warranty.

Peter

[mckrause]

You have a leaking ferrule - probably the one to the transfer line. They re-seal with temperature - that's why you see the drop as the temp program kicks in. Go in and tighten down the ferrules at the injector and the transfer line.

[donnygg]

Thanks everyone for the suggestions! We've just got it fixed. mckrause was spot on - it was a loose transfer line. I tried to tighten the connectors the other day but as it was my first time, I stopped turning as there was a fair bit of resistance. I called the helpline and the engineer assigned the case told me the same thing - tighten half a turn. I was a little scared but click - it could still be tighten! Tuned again and the air peaks and high baseline are gone! Apparently, the joints loosen after shutdown and have to be tighten after restart. Hopefully there isn't any major damage from the leak