SPME Random peaks

[NZBen]

Am a newbie and don't really know what I'm doing.... But trying to teach myself how to GC...

I'm currently Running SPME samples from different plastic building materials and am getting random peaks that I assume aren't real gasses. I'm getting some awful looking peaks with huge trailing edges and some very sharp and tall peaks. It seems to me that the wide peaks with trailing edges are legitimate gasses and the tall peaks are not. What is this artifact from and what could be causing it?

The GC is a 7890A with a 5977A MSD I'm using an HP-5MS column and a CAR/PDMS fiber.

Attached is a snap from a section of the run.

Thanks

https://drive.google.com/file/d/1PHwCY8 ... sp=sharing

[rb6banjo]

When you try to do spectral matching on the various peaks, what do you get? Especially on those blobby-sorts of materials like 4 minutes and 6.2 minutes. The shapes of some of your peaks look good but some not so good. There's just too much going on to type it out on a forum like this.

Learning this stuff on your own will be difficult. Your company should put you in a basic GC class. I'm sure companies like Agilent (your equipment) offer them. it'll be money well spent in the long run.

If you can't do that, have you checked youtube? There are probably GC videos there from which you can learn the basics.

[Gizmo]

If you did not already do this, I'd suggest making sure the fiber was conditioned in the injector for a good half hour before you use it for the first time - and then be sure to run a blank of the fiber to see the artifacts that might still be bleeding off of it.

[NZBen]

Thanks for the thoughts so far, a couple of things I negated to mention...

I conditioned the fiber for an hour at 300C before use, I suspect the horrific tailing peaks are caused by an inlet condition (I think I need to cut and reinstall the column into the inlet). I also ran a condition on the column and a bakeout on the MS. Water shows quite high in the MS (around 18% - other gasses are low, around 0.5-3%)

At the moment I'm not concerned with the tailing peaks, they match the NIST spectra very closely and are compounds that would be expected to be present in the sample. I can fix the long tails later. My prime concern is the sharp peaks. They do not match the NIST library, the number one option is miles off when you compare them side by side, they are also unexpected compounds. I've had an RFPA difficulty error previously, so I suspect the sharp (good looking) peaks are an artifact caused by an electronics problem or chamber pressure (am hoping it is the high water content to blame). I don't have an ion gauge, but the foreline pressure is around 30mTorr. (makes me feel nauseous just saying that - why do they use such horrific units?!?!?) I figure with the foreline being at that level, the chamber pressure should be ok? The pump is a diffusion pump with good clean oil.

[NZBen]

Well looks like it is solved, I don't really understand what happened, but I ran several blank runs and found those sharp spikes kept reducing, then there were none... so no idea. My peaks from my SPME are all looking good with minimal tailing and they all match up great with the spectral libraries.

So yeah - the RFPA fault seemed to disappear my itself too. - I baked it a few more time till that point and the water reduced to 3% and it seemed fine after that. Maybe it was the chamber pressure too high?

Thanks for looking over this thread and providing some insights. Hopefully it can help someone else who comes across a similar issue.