Baseline issue

[sg]

Dear all,

I currently encounter a baseline problem. I use DB-624 column to run a impurity method. The problem is that at the end of the run when the highest temperature is reached (last 4 minutes running), the baseline become very noisy (wavy). I have an impurity elute there, this noisy baseline affect the sensitivity.

So please could you provide some suggestion what is the potential cause for this?

The sample is dissolved in ethanol.
The GC parameters:

Dect: 300 C
Inj: 200
Carrier gas: Helium
Programming : 50 (2) ramp to 260 at 15 C/min and hold 4 minutes.
Split at 20:1
Injection volume: 1uL.

Thank you!

[Peter Apps]

Hi sg

This sounds like column bleed.

You might have oxygen in your carrier gas in which case you need a scrubber in the gas line just upstreamof the GC, and a thorough leak check of all the connections in the GC.

Also, the column might be attacked by water in the samples (ordinary "pure" ethanol contains 4% water). Drying them might be tricky, would it be possible to use another solvent ?

Peter

[chromatographer1]

I agree with Peter that bleed may be the problem. Now, what to do?

Do you condition the column at 260C before cooling the column and running your sample?

Is your column old? How many injections and high temperature sessions has it seen?

You are running your detector at 300°C. This will heat the end of the column (probably already damaged the end of the column within the detector) and conduct heat back up towards the front of the column and can damage it, especially when the oven is at 260C (less cooling of the column below its maximum) and phase can breakdown and bleed off, especially if you have oxygen damage as Peter mentioned as a possible cause.

WHY 300C? Why not 260C or even 270C?

The maximum temperatures recommended by manufacturers are selected because of known limits of thermolytic degradation. Please do not exceed them unless you don't mind a wavy baseline at the maximum temperature.

You might wish to consider using a thinner film column inorder to elute your late impurity at a lower temperature.

best wishes,

Rod

[Bruce Hamilton]

Whilst first possibility is column bleed, that usually should go up and then taper down as single broad peak, and should decrease during a sequence.

My guess would be higher boiling rubbish in your sample, solvent or any derivatisation reagent ( if you use one ) eluting from earlier injections. If you condition the column and then run multiple solvent injections, does the tidal wave pattern get smaller and less frequent?.

I'd change the column to one that was more amenable to high temperatures and see if there are peaks that elute off later than you expect in your samples. I'm not a big fan of running polar columns near their maximum temperature - they tend to die quicker, with lots of blood and gore.

If you don't see any peaks at higher temperatures, then it may be column bleed or sample degradation in the injector, in which case try extending your run for 20-30 minutes at the final temperature to see if they diminish to broad flat waves that are less intrusive.

bruce Hamilton

[sg]

Hi there,
Thank you very much for your reply!
The method I can not change unfortunately. The problem is that our R&D analyst use the same packing column and condition and can obtain very good baseline. I isolate to troubleshoot and seems there is no problem with detector, injector... really column problem.. but not understand why they can get it... so I suggest either we mail them our column and let them run or ask them to send their column for us to trial... to see if it is instrument itself prob..

The gas we use are pure and there is stand way for purification before gas enter GC...

Any thoughts?

[Bruce Hamilton]

Go for swapping the columns, if theirs works OK just don't bother returning it, or only accept a brand new column in exchange .

If you believe it's the instrument, it would be in the injector or detector, and if other columns work OK at the method temperatures, the likelihood is that the column is feeding waves of old muck to your detector. If the problem is the instrument, clean out the injector first, and if the detector is generating waves ( very rare ), it's probably time for a clean as well.

Good luck.

Bruce Hamilton