[Zudon]

Hi guys.  I am currently working on a design project for a xylene plant.  The process I'm using is Toluene disproportionation and operating conditions of my xylene column (to separate xylene from heavier aromatics like TMB), are:

Top Temp.: 123 oC
Bott Temp.: 152.5 oC

No. of stages: 16

I have calculated the number of stages in the rectifying (r) and stripping (s) sections, and this is what I got:

Nr = 11
Ns = 5

(rounded off to nearest whole number)


There are more stages in the rectifying section than in the stripping section.  May I know if this is the normal case in a typical distillation column in a chemical plant?  And if yes, why would it be this way? 

Thank you.

[RealEngineer]

Hi Zudon

The rectifying section merely refers to the stages above the feed (with the stripping section below the feed).  I assume that you are feeding the distillation column at the tray with the closest composition to that of the feed itself - this is the optimum design. 

If so, the location of the feed tray and the number of stages required in the rectifying and stripping sections will depend on the feed composition and the required composition in the bottom and top products.

It would be worth playing around with you design to test this.  I think that I have a simple distillation model which you can borrow, if this is of any use.

Best of luck with your design.

[Zudon]

Hi RealEngineer.

Thanks for the help.  Yes, I would like to borrow your model to test my design.  How do I get it from you?

Thanks again.

[Zudon]

I have another question, this time relating to a Separator that is operating at T=115oC.  I have a stream entering the separator and it has the following components in these compositions:

Benzene 0.0072
Toluene 0.9831
Xylene 0.0096
Ethylbenzene 0.00005

K-values for the components at T=115 deg Celcius are:
Benzene 3.3
Toluene 1.6
Xylene 0.78
Ethylbenzene 0.78



I want to separate Toluene from Xylene and Ethylbenzene, but by doing so I should have Benzene and Toluene exiting the separator in the vapour phase as Benzene is more volatile than Toluene.

I used a Flash Distillation method to calculate how much components would exit in the vapour and liquid streams.  This requires me to guess a L/V value first, then see if the actual L/V value is similar to the one I guessed at first.

However, once i calculated it in Excel, the L/V has to be negative to "converge".  Now, does this mean I have made a serious error in my calculations, or that separation using this method is not possible.  If not possible, does this mean I have to use a distillation column instead?

[RealEngineer]

Hi Zudon

As you guessed, you have either choosen parameters which are not physically possible or made an error in the calculation.  Without seeing the details of your model, I can't be sure which it is!  However, the following are a couple of suggestions.

1.  The relative volatilities show that you will get a decent separation with Benzene / Toluene in the vapour phase and the other components in the liquid.   However, in a single stage flash there will still be a lot of the light components in the liquid and vice versa.  I would suggest that to start with you try a less rigorous separation and try to get that to converge.  As you have guessed, if you want a better separation, you probably need a distillation column as this will give you more theoretical stages.

2.  Trying to converge a solution by guessing L/V is notoriously difficult, as it is unbounded - i.e. it can vary between zero and infinity.  Try guessing L/F or V/F (where F is the Feed) - this can only vary between zero and 1.

If you contact me via my email address (see my profile) I will forward you the simple distillation model. 
 

[Zudon]

Hi RealEngineer.  Thank you so much for your help via email.  I really appreciate it.

I am facing another problem right now, about the choice of column I should design.  Right now, I wish to separate "Xylenes" from a byproduct of the reaction called "Trimethylbenzene".  I have started designing a sieve tray  column for this, but now I heard that another group doing the same project is using a packing column instead so a high purity of Xylene can be produced, since Xylene is my targeted product.

I am unclear on the differences and similarities of the two types of columns, and would like to know which of the two is recommended.

Thanks.

[RealEngineer]

Hi Zudon

I am not sure that you will get a  better quality of product by changing from trays to packing.  The quality of the separation depends on the number of theoretical stages. 

Usually the choice between trays and packing is a matter of economics.  Packing columns tend to be used in small diameter columns while trays are cheaper for big columns.  So distillation columns to refine crude oils are trayed while distillation of fine chemicals is usually done in packed columns.

There are other factors to consider (e.g. packed columns usually have lower pressure drop and can handle lower liquid flows, while trays are easier to clean when distilling fouling liquids).  A decent chemical engineering textbook should discuss this better than I can do.

Regards

RealEngineer