6 replies to this topic

[Geetanjali]

Hello members,

 

Hope you are doing well.

 

I am modelling the separation of n-butane from 1-butene. Because these are close boiling liquids with BP of butane = -1 degC and that of 1-butene = -6.3 degC, it is being advised in the literature to use an Extractive distillation column using polar solvents like acetonitrile or  n-methyl pyrrolidine mixed with water (in the ratio 0.8:0.2). Ideally, the n-butane should come off from top of the ED column and the polar solvent mixed with butene should flow down from the bottom of first column to another column - a stripper- where we get rid of butene from the top and the solvent to be recovered from the bottom (and can be recycled). It should be an easy separation because the BP of butene = -6.3 degC and that of DMF is 202 deg C (and if we use acetonitrile, then it is 87 deg C).

 

I am referring to a patent that describes that the Extractive distillation column should have 70 bubble cap trays with a column diameter of 80 mm.

 

The problem is I am unable to get n-butane coming out from the top of the first column after trying all possibilities. The patent says that butene should come mixed with solvent from the bottom and it should be separated in the second column from solvent. However, in my case, butene is coming out from the top of first column instead of n-butane. 

 

Can you please provide me some advice on this? Thank you

 

I am attaching the patent.  

 

Best,

Geetanjali

Attached Files

•  US Patent 20060096849A1- Butane and Butene separation using a polar extractant.pdf   453.25KB   19 downloads
•  Butane vs butene separation.JPG   30.68KB   6 downloads

[PingPong]

Maybe you are no using the right thermodynamics in your simulation. You need to use the right activity coefficient model, not simply PR or SRK.

 

Attached pdf gives a good example. They however do not use DMF but instead acetonitril (ACN), which is also used in the patent under [0019].

 

First try to reproduce their results with ACN and when successful replace ACN by DMF if that is what you want to use. That may require adjusting other parameters as well such as pressure, temperature, solvent recycle rate, number of trays, et cetera.

 

 Process Simulation of 1-Butene and N-Butane Separation By Extractive Distillation.pdf   535.12KB   31 downloads

[Geetanjali]

Thank you for your insights. I was successful in getting the separations. I used NRTL property method and used Dimethylformamide (DMF) as solvent. I used 2 columns, got butane out from the top from the first column and separated butene (again from top) and solvent (from the bottom) of the second column. It took some iterations and lot of literature searching, but finally, I am happy that it worked. I followed that patent I attached in my first post.

 

 

Thank you for your time

 

Best regards,

Geetanjali

[TWL1139]

Maybe you are no using the right thermodynamics in your simulation. You need to use the right activity coefficient model, not simply PR or SRK.

 

Attached pdf gives a good example. They however do not use DMF but instead acetonitril (ACN), which is also used in the patent under [0019].

 

First try to reproduce their results with ACN and when successful replace ACN by DMF if that is what you want to use. That may require adjusting other parameters as well such as pressure, temperature, solvent recycle rate, number of trays, et cetera.

 

Process Simulation of 1-Butene and N-Butane Separation By Extractive Distillation.pdf

Hi I am working using this pdf as well but when I worked it on Aspen HYSYS it showed that the solvent stream does not have enough information to run the column, is there anything I can do with that?

[Pilesar]

The commercial process using NMP has the butane product leaving the top of the first column (the butene absorber) with the rich solvent leaving the bottom. The butene product leaves the top of the second column (the butene stripper.) The lean NMP solvent leaves the bottom of the butene stripper and is cooled before entering the top of the butene absorber.

  Typical recovery is 99% butenes. The butene product will be 90-99% butenes. The butane product will be 85-95% butanes.

Edited by Pilesar, 24 October 2023 - 12:17 PM.

[Geetanjali]

Thank you so much. This is very useful!

[Hong Kiat]

Thank you for your insights. I was successful in getting the separations. I used NRTL property method and used Dimethylformamide (DMF) as solvent. I used 2 columns, got butane out from the top from the first column and separated butene (again from top) and solvent (from the bottom) of the second column. It took some iterations and lot of literature searching, but finally, I am happy that it worked. I followed that patent I attached in my first post.

 

 

Thank you for your time

 

Best regards,

Geetanjali

 

Hi Geetanjali. I am also working on this simulation, however I can't make my simulation converged. Mind sharing your Aspen HYSYS simulation file? Thanks a lot.