8 replies to this topic

[smartiron]

Hello all,

can somebody help me with an azeotropic distillation?

I need to separate isopropanol from water to get pure substances. I know, from Aspen Properties, this mixture forms an azeotrope with min boiling point at 80,38°C and with 0,675 mol% of isopropanol (hope, its right).
My mixture contains 0,46 mol% isopropanol and a rest is water. I could separate pure water using a simple distillation and I got a new mixture of 0,65 mol% isopropanol and 0,35 mol% water. Thats all, I could do in HYSYS. Now, I dont know what to do. I tried to use a distillation column at 2 atm to break the azeotrope and I used a partial condenser. I got 2 streams with 0,662 and 0,667 mol% of isopropanol, from bottoms Ive got pure water.
What is the next step? Or what method do I have to use?

Thanks..

[luuquocdai]

Hello all,

can somebody help me with an azeotropic distillation?

I need to separate isopropanol from water to get pure substances. I know, from Aspen Properties, this mixture forms an azeotrope with min boiling point at 80,38°C and with 0,675 mol% of isopropanol (hope, its right).
My mixture contains 0,46 mol% isopropanol and a rest is water. I could separate pure water using a simple distillation and I got a new mixture of 0,65 mol% isopropanol and 0,35 mol% water. Thats all, I could do in HYSYS. Now, I dont know what to do. I tried to use a distillation column at 2 atm to break the azeotrope and I used a partial condenser. I got 2 streams with 0,662 and 0,667 mol% of isopropanol, from bottoms Ive got pure water.
What is the next step? Or what method do I have to use?

Thanks..

I think that, you can use special distilations to break azeotropic mixture. We have many special distillatons, please try to research about it.

[jt217]

use a decanter in between the two columns

[smartiron]

Yes, Im going to use a decanter. But seems to be more difficult as I thought

[Andrei]

smartiron

Usually one tower should be enough for the separation.
You have pure water separated at the bottom of your tower and the azeotrope at the top. The beauty of this is the fact that the azeotrope is not miscible. You will get two liquid layers one water and one isopropanol after condensation. The solubility of one component in another one is relatively small and depends on the temperature. The lower the temperature after the condenser the higher purity of the two phases. Isopropanol will be the tower top product. Some of the water will have to be returned in the tower as reflux (usually almost all of it), depending on the tower design you have. If you need advanced water removal from isopropanol, some water desicants are used.
This scheme worked when I used to operate an oxo-alcohols plant.

[smartiron]

Hi Andrei,

can you, please, write more about this process? What do you thing under the temperature after the condenser? Did you mean to cool the distillate from condenser to lower temperature to have two different streams? Or to cool a vapor from a head of column? It is the first time I read about this process.. As I wrote, I wanted to use the third component and a decanter, but it still does not run. (I am doing it in HYSYS)

Thanks a lot..

[Andrei]

smartiron,

To be able to simulate two liquid phases at the top of the tower you have to use a two liquid phase thermodynamic system. I never done the simulation in Hysys, but in PRO/II NRTL was working fine, UNIQUAC was also giving good results. And of course you have to make sure that your simulator has the required interaction parameters and they are describing the two liquid phase behavior. This should work in Hysys also with suitable thermodynamics.
Regarding the cooling, usually the condenser temperature is selected some 5-10 degrees below dew point at condenser inlet from energy conservation considerations. In this case you may choose a lower condenser temperature to ensure a net separation of the two liquid phases, the lower the better.
Your condensate drum needs to do the two liquid phases separation, return one phase (water) as reflux, and extract the other one (alcohol) as top product.
You don't have to use a third component for this distillation, that is done only for miscible azeotropes (like ethanol-water), this is not miscible.
I've seen this working in the plant. If Hysys fails to show this behavior it means that the thermodynamic system is not the right one. I warn you, it is very difficult to bring the tower into convergence, you have to use special convergence algorithms, and usually inside-out fails.
Good Luck

[smartiron]

Im still not able to simulate this process which you have described to me.. There may be a problem with thermodynamic parametres.
I could only get two liquid phases with cyclohexane as a third component (entrainer). And it was easy to separate these two phases to aquatic and organic streams, after all. Since I coulnd find a LLE estimation of binary parametres, I decided to use a standart method with decanter. But another new problem has occurred. I would like to reflux the organic phase from decanter to a distillation column. And I cant set this in Hysys.. Does anybody have any skills with this connection? Thanks

[Andrei]

It's OK, it can work also the way you described, but you need a second tower to separate cyclehexane and recycle it.
With the second problem I cannot help you without seeing the flowsheet.
Why don't you send your flowsheet to Aspentech support? I am sure they will help you.