Chromatography Forum

Hi,

According to what I've learned about LC pump, even if I run out of solvant, my column will never dry because pump will lost prime and will not be able to pump air into the system (correct me if I'm wrong).

But, if for example the system is pumping 50%A and 50%B and only the A line goes dry, is there a danger for introduction of air bubbles in the column before pump stalling?
Or will the system be pumping the equivalent of 100%B on column without introduction of air due to absence of input from A?

Something related happened to us this weekend. The system was idling at 95%Water : 5%Methanol on a C18 column and we ran out of A. On monday morning, the error message was SYSTEM OVERPRESSURE and since that time the pressure of the column is higher than it was on last friday and we get overpressure message during gradient run. I know that max viscosity for a MeOH/H2O mix is somewhere around 50%, but this method was running well before. It looks like there is an air bubble trapped on the column and up until now, circulation of MetOH had'nt helped a lot.

Does anyone have a suggestion (and an answer to my technical interrogation)?

Thanks

It is not an airbubble that is the problem. My bet is that the issue is debris from pump seals that clogged the inlet frit of your column. I would attempt to replace the inlet frit.

Uwe,

Thanks for your reply,

Laste nignt, methanol was passed up flow, precolumn cartridge was changed, then the pressure decreased, but not as low as it was before. So it seems consistent with the clugging theory.

I just would like to figure out how this happened and how to avoid it, so:

How do you come to this conclusion? Are those debris more likely to come out from pump seals when a line is going dry? or is it not related to my dry line incident?

We have an inlet filter between the pump and the injector. To reach the column, the pump seal debris may have gone through it...

May I conclude that it's not possible to pump air into column even if there is still liquid remaining in one of the lines?

Martin