I'm at a new lab that needs to test fuels/oils/solvents - sometimes with a lot of water - for PCBs.

The many years old (and many years unused) SOP I was given uses a column to clean up the samples for GC-ECD.

Add 1/2" silica gel
Add 3 inches Florisil
Add 1/2" silica gel
Add 3 inches acid-treated silica gel
Add 2 inches sodium sulfate

Does that seem overly complicated? For this application, is silica gel going to do anything that florisil isn't? Is there any reason to sandwich florisil between silica gel layers?

I intend to try a few samples using the above method and something simpler (maybe just omitting the silica gel). But this facility gets a wide variety of samples, and I'm not yet familiar with everything I might see. I'd rather not eliminate part of a procedure without understanding its intended purpose.