Chromatography Forum

Hi all,
Im doing GC analysis with dirty samples. As a result one problem I encounter is the carry over contamination from a previous injection. to remedy this I run a "purging" method which involved a high split ratio to purge the inlet. Then I injected a blank sample injection just to ensure the purging was sufficient and I always rinse the sample syringe 3 times with wash solvent before and after a sample injection

despite this I still on some occasion encounter carry over despite the purging method. whats interesting is the carry over contaminants tend to have the same retention time over a series of run just wondering if anyone have experience this before? and whats the best remedy for this.

If you are running samples whose matrix is similar then it is not unexpected that the carryover peaks will be due to the same substances, and so will have the same retention times. It is possible, nonetheless, that at least some of what looks like carryover is due to septum bleed, bleed from any packing that you have in the inlet, or to contaminants in the carrier gas that are accumulating at the top of the column.

Since running the purge method presumably cuts your sample throughput in half, in your place I would be looking very seriously at better sample cleanup.

Three syringe rinses is the minimum that you should use - try increasing the rinse cylses, and using an extra solvent different to what the samples are dissolved in.

Peter

In my experience it turned out that one can not really clean syringes without removing the piston and washing both parts separatly.

I've found that blank injections can be misleading (if you mean starting the GC without injecting anything). These can look clean and subsequent "real" runs still display evidence of contamination. You're better off shooting a blank solvent into the inlet for the blank runs. This was shown to me and explained as "you need to put something in the inlet to get a thermal load on it". That may not be the real explanation, but this approach seems to work better.

Its a good idea to clean or replace the split vent line (not just the trap) if you suspect you have flashed sample up into it.