Chromatography Forum

Greetings and thank you in advance. I am relatively new to GC-MS on my own, so any help is appreciated.

I have a Perkin Elmer GC-MS (Clarus 580 - SQ 8 S) with Turbomass software that was setup in January 2016 by a rep, but never tested. After correcting some cables, instrument and PC are interfacing and operating ... but results are poor (see picture links below). Eventually, will be studying tea extracts, so PI wants to use MeOH as a sample solvent (seem reasonable?). Injecting 1uL into a 200C injector, no split on injection, carrier gas at 2mL/min, split flow 75mL/min, running on a PE Elite 17-MS column (rated for operation between 40C - 300/320C). Tunes and calibrates well, system is leak free. As funds are limited, I certainly don't want to waste money while trying to resolve the issue ... as well as not introducing additional problems.


Screen Shots:
1) https://drive.google.com/open?id=0B8Z18 ... E9oaGpyLVU
Blank MeOH Sample. (Also injected an air sample with similar result)
2) https://drive.google.com/open?id=0B8Z18 ... WRqV1A4NHM
Used a sample of Propyl 4-hydroxy-benzoate, provided by PI
Please note: Top portion of each picture is scaled to column saturation at 1e10, and bottom portion is scaled to mid detection levels 1e8.

Primary problems:
1 & 2) RELATIVELY HIGH BASELINE & BLEEDING. Was told limit of detection is around 1e7, and saturation around 1e10. While temperature is below 140C, baseline itself is adequate, but it climbs up to 1e8 and 1e9 by 210C, and bleeds at higher temperatures (the blank shown, final temp was 275C; the sample shown, final temp was 200C). To get adequate signal in this scenario, I need to keep temperatures quite modest, and have very strong signals bumping up near saturation ... anything at e8 and lower has interference.

3) RANDOM PEAKS. Get constant blips in the baseline. While they become more frequent and more intense at higher temperatures, they are present throughout. Always similar masses: (73, 147, 221, 281, 355) at the lower temperatures, and (150, 253, 315, 393, 451) at higher temperatures. Library search usually matches to variants of siloxanes.


My interpretation is that all of the problems are all related to a column that has gone bad and is degrading. Constant shedding gives random peaks, poor absorption onto the column, sample required for acceptable signal:noise nears column saturation and causes broader peaks and tailing, and column degrades above 240C to cause bleeding. Is this reasonable? Any other obvious cause I should investigate? And solution is just to replace the column? Is there a way to recondition it to stop the shedding and save costs? Acceptable to operate under the current conditions (ramps under 20C/min, max temp 215C)?

Trying to understand why the column would've gone bad ... It was sitting around for almost 7 months with just the occasional injection. Age and minimal maintenance kill it? Is there a solvent/column compatibility issue? If column is rated to 320C, why does it bleed at 215C and sheds throughout? Ramps were within limits (5 - 20C/min) and max temperature used was at least 25C under column max.


Any advice ...? Just spend the 800 necessary for the new column? :/ Thank you again

Columns are generally pretty rugged. Generally, they only fail when they are held at high temperature AND exposed to oxygen (from the air most of the time). I have to say that I'm an Agilent guy so I don't know a lot about the "ins-and-outs" of PE equipment. I have to say that I tried the 50:50 phase (50% phenyl:50% dimethyl) from another vendor and it bleeds like crazy (I've only used it on a flame detector)! I can seemingly never get it to stop. After installation, it's really ridiculous but even after it's been installed, you arrive in the morning and raise the temperature to the limit and that first time up is pretty bad. I tend not to use this column because of that. I would rather just use a wax. For tea extracts, you might be able to get by with a 5-phase (5% phenyl:95% dimethyl). None of mine bleed anything like that 50:50 phase.

Those masses you quote are often seen coming from siloxanes. It could be column bleed, could be an inlet septum problem. Do you see the same chromatograms if you put the GCMS though a run cycle without an injection? To do this, set the instrument up for an injection (computer to collect data as well) but just start the instrument - don't actually inject. Have you tried to set the temperature of your column near the isothermal limit and just let it bake out for a while (overnight maybe)? I assume that you have electronic pressure control. If yes, I would increase the column temperature to the high value AND double the head pressure on the column. Force stuff out faster.

Propyl-4-hydroxybenzoate is probably not the compound you want to inject to assess how well your system is performing during troubleshooting. Phenols can be difficult to chromatograph. It may be fine but I would choose something simpler. Choose something that you know will interact well with your stationary phase.

Good luck. I think your problems will go away if you change columns. My gut feeling is that the 50:50 phase is your problem.

Oxygen and water in the carrier gas kill columns quickly and they bleed a lot while they die, so you need an oxygen and moisture scrubber on your carrier gas line immediately upstream of the GC and a thorough check of all connections for leaks that will let air diffuse into the carrier gas. Was the carrier gas always flowing while the machine was idle ?

Peter

The column need to be conditioned (first disconnect it from the detector inlet) and injector liner should be replaced. Peaks are good in shape, peaks from previous injection are observed as well. What kind of samples were analyzed before high background appeared?