Column Contamination

[mdewald]

Hopefully someone can help me with this. I'm running EPA Method 515.4 on a DB-5MS UI column as the primary and a Rtx-50 as the confirmation column. The DB column is 60 meters. I'm seeing contamination peaks show up after a month, sometimes longer, in the primary column that I can't get rid of. Baking off helps only slightly as does port maintenance. The confirmation column is helped by the baking off. Has anyone else seen this problem? Could it be the column its self breaking down from the injections? I would think that if it was something accumulating from the materials used in the extraction that it would mess up the Rtx-50 column in the same manner i.e. not be able to bake off the contaminant. Any help would be appreciated!

[Gerhard Kratz]

Maybe it will help when you cut 15cm of your DB capillary. Otherwise try a new capillary. How many injections your DB capillary saw?

[mdewald]

I've cut about a meter off of the DB column with no success. The column can see as few as 100 injections before things start to break down.

[Peter Apps]

Do you have oxygen and moisture scrubbers on your carrier gas or if you have them are they exhausted ?

Baking columns quite often just kills them more quickly.

Peter

[gcguy]

You could try rinsing with an appropriate solvent(s). There are kits available for this, I have had quite a bit of luck getting contaminated columns back in use after rinsing them.

GCguy

[hossam kamal]

I think it may be column bleeding . Try another column with the same packing and see if the same problem still exist

[rb6banjo]

Please excuse my ignorance but I don't know this method. Is that Rtx-50 used in every analysis? I have an Rtx-50 and I don't like it because it bleeds like crazy. I can bake it out, cool it to 40 °C, then run a blank baseline (no injection) and still get a ridiculous amount of bleed - and it's not just that nice sloping bleed, it's discrete peaks. I never use it because of this problem.

[James_Ball]

Are you doing dual injections on two separate ports or splitting one injection between two columns?

Also wondering if you are getting some septa particles down into the column, those will give some strange peaks that can't be baked out.