4 replies to this topic

[Sarah_Process_engineer]

I am working on a GTL process modeling and simulation using the tri - reforming process. No I need to simulate the Fischer Tropsch reactor, but I blocked when I wanted to enter the reactions that can take place ( CO + H2 --> (-CH2-CH2-)n + H2O ). 
It would be very nice of you if someone could help me solve this problem since It's my first time dealing with reactors in Hysys Aspen. 

Thank you very much

[Bobby Strain]

Not many of us would undertake this chore. I suggest that you use a simple reactor model and define a set of reactions that produce the products typically formed. With this, you can then proceed to model the downstream portion of the process.

 

Bobby

[abubakarce19]

use gibbs reactor and check products at your reaction T & P. Than add reactions involving those products in conversion reactor, iff you know desired conversion %ages. Otherwise use equilibrium reactor.

[shantanuk100]

Hello Sarah,

 

For Fischer Tropsch Process, it is a patented one and there are a lot of uncertain variables such as reaction kinetics, constants, catalysts etc., that can't be found without prior consultation of the relevant parties if you wish to accurately model the reaction and process. That said, if you wish to do a preliminary simulation of the process, then that is possible.

For polymer reactions it is better to use Aspen Polymer Plus.

If you have the POLYPCSF Package then you can use that in hysys.

But the 1st alternative is preferable.

Whether in Hysys or in Polymer plus, the steps are more or less as follows in the PDF Document I have attached below.

Please find it below.

Also refer to these links for any further clarifications.

 

http://www.owlnet.ri.../reactions.html

http://www.owlnet.ri...ysys/gibbs.html

http://www.owlnet.ri...onvreactor.html

http://www.owlnet.ri.../hysys/pfex.htm

 

 

Thanks,

Shantanu

 

Attached Files

•  Reactor Simulation - Hysys.pdf   802.31KB   79 downloads

Edited by shantanuk100, 12 November 2015 - 02:16 AM.

[PingPong]

I agree with Bobby.

 

You are dealing here with proprietary technology that is owned by only a few companies like Shell and Sasol. Only a few engineers are able to model such FT reactors reliably. All you (and I and Bobby) can do is a rough approximation of the product yields based on what can be found in the open literature, assuming you can find anything useful.

Note however that FT can be designed to produce many different products. Sasol in South Africa makes different products than the Oryx plant in Qatar that also uses Sasol Technology. The latter produces mainly middle distillates. The Pearl plant in Qatar uses Shell technology to also produce middle distillates.

So you have to decide first what kind of plant you want to model.

 

You should be familiar with the Anderson-Schulz-Flory (ASF) distribution. FT product slate is determined by the chain growth probability α (alpha), which depends on operating conditions, H₂/CO ratio, catalyst type, reactor design, and what else.

Shell uses an α that is very high, somewhere between 0.9 and 0.95 and therefor mainly waxes are produced. These waxes are further processed in a Hydrocracker to produce the desired middle distillates and base oils (for lube oil manufacturing). Oryz also produces wax that is processed in a Hydrocracker.

Sasol in South Africa makes lighter products of all kinds, by using a much lower α of maybe 0.7