CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Defining a non-granular phase (

Tom September 10, 2003 05:04

Defining a non-granular phase

I am trying to use the Eulerian Model (Fluent 6.1.22). I am trying to model a three phase system. The primary phase is caustic soda, the secondary phase is a granular particle and I would like the third phase to be water. My question is, how define this third phase as a liquid? When selecting the phase I am asked to input a diameter! I am assuming this is for something like an air bubbles, however, I would just like two liquid phases that will end up mixing together.

Any suggestion please.


ap September 11, 2003 07:57

Re: Defining a non-granular phase
Your system is a complex, because you have a solution of NaOH in water as primary phase, a secondary granular phase and water as third phase, which is soluble

I don't know your operating conditions, however, I suppose you're far from the limit of solubility of the NaOH. If not, you should account for that.

Eulerian model considers all phases as interpenetrating continua, which occupy a volume in the domain proportional to their volume fractions.

I'm not sure you can model the mixing of water and NaOH solution using the Eulerian Model, in which phases are kept separated and their interactions are accounted using a momentum exchange term.



Tom September 11, 2003 09:12

Re: Defining a non-granular phase
Thanks for your reply. I have tried using the eulerian model, however, I could not get it to work. I'm now having a go with the mixing model.

I still dont understand how Fluent deals with two fluids, as I am asked to input a diameter? I'm sure this should be possible.

ap September 11, 2003 15:12

Re: Defining a non-granular phase
In the Eulerian Model all phase, also granular ones, are considered as fluid.

You usually have a primary fluid phase, and a number of secondary fluid phases.

For granular phases, the meaning of diameter is evident because it is the particle diameter.

For secondary fluid phases, the meaning of diameter is somewhat less evident.

In a bubble column (liquid-gas) it's the diameter of a bubble of gas. If the primary phase is a gas, and the secondary is a liquid, the diameter is the diameter of liquid drops. Look the tutorial case "Using the Mixture and Eulerian multiphase models" for this last case.

If you have two or more liquids, which are non soluble one into the other, like oil and water, you can think that the secondary phase forms "drops" in the primary one, so the diameter is the drop diameter.

This is why I don't think you can model your system using Eulerian model: in your system water and NaOH solution mix up giving just a more diluite solution which is a single phase.

Anyway, I'm looking for more information.

I'll tell you if I found something useful.

Hi :)


Tom September 12, 2003 03:35

Re: Defining a non-granular phase
Cheers ap, much appreciated.


thomas September 18, 2003 07:05

Re: Defining a non-granular phase
Hi, Let me express an idea to your problem. You should set your problem not as a 3 phases case but as a 2 phases including a granular phase and a multispecies liquid phase ( 2 species : water and NaOH ). Eulerian multispecies solver is not available yet (it will be very soon). For now what I could propose to you is to declare a 2 phases case (granular phase + water for example) and then use UDS to tranport NaOH as 2nd species in the liquid phase. Transport scalar equation also allows to simulate chemical reaction with NaOH. This is clearly explained in the Fluent documentation.

It also exist another solution using a pre-process step to simulate mulspecies gas combustion but I am not familiar with it.

Let me know if this help thomas

Tom September 18, 2003 10:34

Re: Defining a non-granular phase

That sound like an interesting workaround to the problem. Can you give me a little bit more direction on how I go about setting this up (macros to use, procedure...etc...)

I like the idea especially because of the addition of the chemical reaction. This is an important stage and I would like to add it if possible.

Any advice would be much appriciated. Regards. Tom

thomas September 19, 2003 10:54

Re: Defining a non-granular phase
Hey man, Okay I will try to bring you some further informations. First do not make the confusion between multiphase and multispecies. Your case is a 2 phases flow( solid liquid ) with 2 liquid species ( H2O and NaOH ). There I think you have 2 possibilities: Mixture model ( I do not know if there is a granular flow option ) and the Eulerian ( with for sure a Granular flow option ). The multispecies model is not available for the eulerian solver and I think it does for the mixture model : But in both case this is not a problem !

Using the mixture model : Declare 2 phases : the solid one and the water. Then in Define>Models>species set species transport with species define in the material panel. Normally in that case the work is done. Concerning an eventual reaction, Either you set you Reaction using Fluent species reaction rate model either you implement them by source term ( macro Define_source ).

Using the Eulerian model : Declare the granular (solid) phase and the water phase. Then as I said before there is no multispecies model for the fully eulerian Model ( Schedule for beginning of 2004 i think ). Thus your have to create a User Define Scalar (UDS)to transport 'according to me' the NaOH volume fraction. The reaction has to be define using the macro Define_source.

WARNING: What I said before might include mistakes that certainly some people will correct after reading. The best advice I could give you is too contact the Fluent hot-line of your country. But I am positive on the chance of success of this method, It does work for combustion so it should not be a problem for chemical process.

Hope this will not confuse your work progress, regards, thomas.

Ben September 19, 2003 11:52

Re: Defining a non-granular phase
Thank you very much for your explanation. I now have a better understanding of your idea.

I shall give it a go and let you know what happens.

Many thanks again.


All times are GMT -4. The time now is 18:13.