Solvent Loss - N2 Purge

[finechem]

I have a 10,000 gal batch distillation pot with 4 receivers operating 24/7. Currently I have an N2 purge on pot and all receivers. When fractionating (THF @ atmospheric pressure) our solvent losses are enormous (condenser discharge 65F) - no secondary. All vent lines tie into a vent header and knock out pot. I am interested in installing pressure control on this system to pad vessel - solvent losses would then be limited to expansion and displacement. My plan for this is to remove N2 purge from equipment and install an N2 control valve and vent control valve on vent from KO pot. Ideally I would like to control from around 2 in.w.c. to 8 in.w.c. (N2 closes @ 2 and vent begins to open @ 8) - any experience with this? Is this tight control range realistic? Any comments would be appreciated. Thanks.

 

[Compositepro]

That will not work. In distillation the pressure has to be kept very constant. You need to purge at start-up and then switch to precise back-pressure control using nitrogen after the condenser.

 

[StoneCold]

You might be able to just put flow control on your N2 pad system that you have. Reduce the N2 pad flowrate to something you are more comfortable with.
Personally I would purge the system with N2 at startup and then shut it off. (This does require a pressure tight system) Then I would just sweep the vents of the vessels to keep other material in the vent headers from entering. This would stop a large amount of N2 from blinding off the condenser as well.

Just my two cents.

Regards.
StoneCold.

 

[finechem]

Thanks - you are correct that the volumetric flow will be high at times - to elaborate I will have a small control valve for fine control and a larger valve for higher volume operating on a split range pressure controller. I have included a dead band, although I can set to zero only I believe the controller may go hunting. If I am discharging or cooling then I expect pressure will be at lower set point (N2 into system) and when filling or boiling up I would expect pressure to be hovering around the higher set point (probably a slight venting condition). The differential pressure as a result of deadband will not affect the column as the changes will not be sudden while at steady state. Really it boils down to instrumentation and can I get good control/tuning out of this type system. Thanks for your reply - let me know what your thoughts are.