No peaks on chromatogram-already checked common causes-help!

[cawarhur]

I'll be as succinct as possible:

Our setup:

Agilent 7890A with Gerstel Preparative Fraction Collector, Cooled Injection System 4, and MPS autosampler.
30m x 530um x .25mm Rtx-5

Samples:

Hydrocarbons C5-C12 (and possibly higher)

Problem:

No peaks on chromatogram afer installation of 530um ID column (besides one impossibly small and early-eluting peak -- ~20 picoamps before .5 min; solvent is DCM)

We've checked all of the common causes and a few others have been considered as well:

-Able to meet and hold all pressure set-points at inlet and detector (FID) (no leaks)
-FID is lit and set as signal 1
-Autosampler syringe is drawing sample

We've run split and splitless, different sample types, different methods, different injection speeds and volumes...we're running out of ideas.

Any insight?

[mckrause]

Did you check your column flow? If you're getting column flow then you should be getting a peak for the solvent (DCM responds relatively well in a FID). If your FID is lighted, and you're getting signal at FID1 (check the signal level - should be 30-50) then the only other possibility is that it is not coming through the column.

If no signal at signal 1 (completely flat baseline, no noise) then you probably have a dead electrometer.

[Peter Apps]

What carrier gas at what flow are you using ? It is possible that the extra flow through the megabore column is upsetting the flame. You may need to adjust fuel or make up flows to compensate.

There may also be problems with the fatter column blocking the flame tip.

Peter

[DR]

Injecting into the same port that the column is connected to?
(obvious, but it is a mistake that has been made in the past...)

[cawarhur]

@mckrause: We do have column flow and we do have a signal at FID1.

@Peter Apps: We're using Helium. I'm going to install a smaller bore column today and see if it is a problem with the instrument itself or if it is due to some issue created by the mega-bore column. I think you are probably on the right track here....

@DR: Yes, thankfully.

[AICMM]

I know this sounds odd but a long-ago instrument guru told me this and it has held up well, have you tried an injection by hand? If that does not work, the next thing I would do is to pull the megabore out and put some DCM straight down the throat and hook it back up. If peaks, you know the back 1/2 is good, if no peaks, then you can start with the back 1/2 and work your way forward.

Best regards,

AICMM

[LCbob]

Hi

As Peter said, its probably the jet. Did you have trouble lighting the flame, the 0.53mm id columns are rather snug in cap jets.

[cawarhur]

The FID lit without a hitch, really..

[DR]

FID still passes the mirror test? Lighting and staying lit are (sometimes) 2 different things.

[aldehyde]

Take the end of the column out of the FID and stick it in a vial of hexane or methanol and check for flow (could be a break in the column).

Always do a manual injection when dealing with stuff like this, I've seen syringes that looked like they were drawing sample but they actually weren't. And if the sample volume is too low in the vial it might not be picking up any sample.

Also, turn off your FID gasses and then set the hydrogen flow to 75 mL check the make up flow (which should be off) and see if the 'actual' reading increases. If there is a clog in the jet hydrogen will hit the clog and then flow backwards through the make up line and be detected in the pressure controller as make up flow.

You may also want to remove the jet and column and shine a flash light from below to make sure a big chunk of graphite isn't stuck in there. You could be clogging your column when you stick it through a piece of graphite as it goes up into the jet.

[Johnny Rod]

Change the column and seehow you get on. I'd reckon either the column iteslef is blocked or you've rammed a bit of ferrule into the FID (probably the jet). They'll light even when partly blocked. It's worth pulling it out and checking it looks good, or trying a substitute.

[larkl]

I've seen a service tech put a drop of something like acetone right down into the FID chminey to see if the FID is responding. I was amazed.