No replies to this topic

[Desmith475]

Hello, I'm working on my capstone project and it involves separating pyridine and methylpyridine out of the waste water of a pyridine synthesis process, and I'm having some struggles with simulating it. The goal of the project is to bypass the pyridine-water azeotrope to get 99.75% purity pyridine and it has multiple impurities that also need to be dealt with. 

 

Going off of a paper which proposes an extraction and stripping process for separating pyridine and water, they used n-propyl formate, and an expired patent gave a clue that DMSO might also be a good solvent for this purpose. Unfortunately it is missing information and can't calculate UNIFAC LLE binary coefficients for n-propyl formate, but it can for DMSO. I looked in the "edit properties" menu of the fluid package tab and saw that a few components were missing variables a, b, and c under the tab "gibbs", which seem to be coefficients for a polynomial. Activity coefficient models, as far as I know, exist to model excess gibbs energy so these missing numbers might help me get a better working simulation. I googled around for info along the lines of "gibbs energy polynomial" and "gibbs coefficients" and couldn't find anything that came close to what I was looking for. Is there any literature out there for this specifically I could track down?

 

Am I just on a wild goose chase? I feel like using simpler models wouldn't be good work for such a highly non-ideal system as this but it may be necessary. Any suggestions for how I might bypass this issue if I can't directly rectify it would be appreciated.

Edited by Desmith475, Today, 06:29 PM.