Chromatography Forum

Hi,

I have lurked around these forums for a while but never posted. I did do a search and read as many threads as I could but I can't seem to find info on this problem. I am using a hypercarb column, heated at 60C. Mobile is water and 50% ACN. I am separating oligosaccharides. I had a method that was working beautifully to separate my substrate from my products. Anyway, I was sharing my HPLC with another lab and the other guy (I swear) took my column off to put his on. He left it in the heater, with no column plugs. I thought that the column would be fine given the nature of the packing, but now I am getting very little or no retention. Any thoughts?

Thanks!

I assume that the aswer is, Yes!
I would not expect that an open column held at 60 °C would show perfect chromatograms.

I wasn't expecting perfection, but hoped for usability. I guess I figured that since you can buy almost the same packing in a dry SPE cartridge that it might be ok.

The dry SPE cartridges are typically conditioned with methanol then flushed with water prior to use.

The other possibility is that the packing reacted with the residual solvent when it dried out.

I would try washing the column with methanol or acetonitrile (or another organic solvent) and trying it again. I note this is porous graphite (not C18) so it may not run properly again.

Off topic, I've heard that columns are ruined if they are allowed to dry out, but SFC columns use carbon dioxide as one of the solvents. Removing those columns can let the carbon dioxide escape- hasn't the column then dried out? SFC columns still seem to run fine when used again. Any thoughts on this? Thanks!

Thanks for the suggestion. I did almost exactly that last night. I ran acetonitrile for about 3h, then H2O for about 6h then tried some standards and all appears to be ok for the sugars I am using. The retention times are about as consistent as they were in the past and my sugar appears to be totally retained (collected flow-through doesn't show anything by MS). As per usual the column does not seem to tolerate relatively fast gradients (changes of more than 1%/min do bad things to the baseline), but that was true before this happened.