Oxygen headspace method driving me crazy...help plz

[Dr-mostafa_omar]

hi all... i'm an analyst with a relatively new experience in GC...i have a method for the analysis of oxygen in the headspace of vials and ampoules...i have agilent 7890A using a molecular seive column 30m x 320mm x 12um...my method uses helium as carrier gas at a flow of 1ml/min with inlet temp of 250C witha a split ratio of 4:1 and the sample is injected manually, column oven at 60C, TCD at 300C...everything was going just perfect...an example run is below:

today i had some samples to analyze, after conditioning the column i injected a sample and i had the following chromatogram:

i tried many injections with the same result, i removed the column and cut from its two ends and re installed it and no change occured, i changed the inlet septum with no use...
please help me what could be the possible cause of this change in one day and what can i do?

[tigerk2001]

Dear Dr.Omar,
It seems by removing the column out of gas and cutting both extremities, you contaminated the column with air and specially moisture. Some moisture is now traps in the column and it is why your system baseline is so noisy.
I would suggest to regenerate(bakeout) the column at 300Celcius on 12 hours period with purified helium as purging gas at a minimum of 10cc/min.
Note : make sure you didn't create leaks on the fittings you played with. Loose fittings or bad installed fittings could also cause such bad reading.

I already have seen such situations with TCD.

regards
Dany

[chromatographer1]

I would guess that your syringe is blocked, or you have a serious leak around your injector.

In any case your flow rate is incorrect, or you are not injecting any sample onto the column due to a blocked syringe.

Since your detector 'zero' is higher now than before you probably have a leak in your injector somewhere causing contaminated carrier gas at the detector.

Trouble shoot the hardware and you should find the problem.

best wishes,

Rodney George
CONSULTANT

[Dr-mostafa_omar]

Thanks alot for your replies...i will try the approaches you suggested