no peaks showing in GC Chromatogram

[Blakey]

When analyzing samples on an Agilent GC using Chemstation software some injections will show no signal except a very small (~300 on the y axis). The next injection will have a normal chromatogram and if we re-inject the same vial it will also have a normal chromatogram. The syringe, liner, septum and column have been changed and the problem still occurs. The more samples the instrument injects the more often it happens (the frequency will increase).

[KM-USA]

So this issue only manifests with the autosampler, and not if you run such sequence using manual injection?

Then sounds like an autosampler issue, and if it appears to get better sometimes after "no peaks", seems to indicate a plugged syringe needle.

Are you using a needle wash either pre-or post injection? Repeated pumping and/or rinse steps?

What's the liquid level in the autosampler vial? Should be like one half to two-thirds the volume of the vial, fully filled can cause a vacuum effect.

[Blakey]

We have changed the syringe and still having the same problem, the vials are 2/3 full and a methanol rinse is done pre and post (2 before and 2 after each sample injection).

[Bigbear]

Make sure the rinse solvent works with your sample solvent. I had issues with methanol when iusing BSTFA as a sample solvent.

[MSCHemist]

Never use a protic solvent as a rinse with a silyating reagent. Use acetonitrile, acetone, MeCl2/Chloroform, pyridine (viscous need a viscocity delay), MEK, ethyl acetate, or a hydrocarbon (hexane, hepatane, pentane etc). Silylating reagents and the analyte products are hydrolyzed by water or any protic solvent (alcohols or amines)

[MSCHemist]

That reminds me are your samples very viscous. You might need to enable a viscocity delay or the injection could be very noreproducible due to insufficient time to draw in sample or getting air in. Also make sure there are enough sample strokes (I usually have 5 or 6).

[Peter Apps]

Hi Blakey

It would be really, really helpful if you told us what it is you are analysing, and the conditions under which you are running - flow rates, temperatures, injection volume, inlet liner type, column, sample solvent etc etc etc.

Otherwise we have to troubleshoot by guesswork, which is slow and unproductive.

Peter