CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Species source and sink (http://www.cfd-online.com/Forums/fluent/79499-species-source-sink.html)

 froniawi August 25, 2010 12:17

Species source and sink

Hi all,

My problem is concerning a sphere in a closed box and a fluid (initially of species A) flowing in between them. Now, I would like the sphere to take up one species (A) and excrete another (B) in a way that the mass-fluxes scale by A_in:B_out=2:1.

As you can see, there is a difference in mass-flux. Theoretically, the difference in mass is taken up by the sphere, which would thus grow in mass and volume. This is not yet implemented, since: it is not essential to me (except, maybe for mass-balance) and I have no idea how to model it.

- Is mass-balance conserved in my actual situation? So, without the difference in mass taken up by the sphere. (In my opinion: No)
- If not, is there a way to model this? Solutions involving a static sphere as well as a growing sphere are welcome.

By the way. Cheating with the densities of the species is not possible, since this problem involves buoyancy.

Best regards

 xrs333 August 25, 2010 12:42

Hi, Froniawi,
IMHO, CFD cannot simulate a fiction. How does the transportation of species A and B take place? Would you please describe the real world phenomenon of your problem?

 Allan Walsh August 25, 2010 13:53

If I understand your question correctly, there is a simple solution in Fluent. Define the volume as a separate fluid zone (i.e. in Gambit), specify that you want to include source terms, and specify the mass sink. If A and B have different properties, then you can specify the species and have a sink of A and source of B.
The means by which you implement is a bit different between Fluent 6 and Fluent 12, which you can look up in help. Good luck.

 froniawi August 25, 2010 17:23

First of all, thanks to the both of you! I really appreciate your help.

@xrs333: As far as I know my research is not classified, but I will keep it a little vague to be sure. Sorry for that.
Imagine me modeling a simplification of a static living being (the sphere) consuming species A and excreting species B. Some of species A is used for self-maintenance and growth. Please note that I am only interested in the resulting flow surrounding the living being (the sphere).

@Allen_Walsh: That solution also came to my mind. It is indeed possible to make a model like that in Fluent and I know how. However, what I am wondering/asking is: Will the mass-balance be satisfied and will the results be "true"? The problem I have with that solution is as follows:
Lets make another simplification and assume that species A and B have the same density. Now. The sphere is consuming two mass-amounts of A and excreting one mass-amount of B (which is equal to one mass-amount of A). This means that one mass-amount of A just vanishes. As a result, the surroundings of the sphere is then decreasing in mass and constant in volume, which means that the density is decreasing while my model is assumed to be incompressible. Hopefully you can prove me wrong!

Alternative solutions I have been thinking about:
- The most straightforward solution: Modeling the growth of the sphere, for instance by remeshing the model at every timestep. Unfortunatly I have no idea how to implement this and it will probably be extremely time consuming.
- Modeling a normal sphere and somewhere far away from it a fictional anti sphere consuming and excreting the exact opposite. This seems like a reasonable solution, but: On the short run, it will require me to investigate the interactions between the normal sphere and anti sphere. On the long run, this solution will become impossible as the model becomes more and more complex.
- Also mesh the inside of the sphere, fill it with species B and apply mass-fluxes and/or a porous zone on the boundary to regulate the mass-fluxes. This also seems like a reasonable solution, but: On the short run, it will only work for short simulations. On the long run, this solution will also become impossible as the model becomes more and more complex.
- Cheating with the densities to make the model mass-conserving. The problem is, my model involves buoyancy.
- Decoupling the momentum and species equations with use of the Boussinesq-approximation? First solving the momentum equations with Boussinesq approximation, then solve the species equations. This is just a crazy idea which is probably not possible, but I believe you should always say what is on your mind during a brainstorm-session. ;)

 byronsmith August 30, 2010 07:49

from what you are saying it appears to me that the sphere is in a mixture of A and B and it consumes A and excretes B at a ratio. if that is the case i suggest you to create a fluid zone around the surface of the sphere. you will have to write a UDF for the rate of consumption of A and excretion of B and apply them as sink and source at this zone. as far as tracking the increasing particle size you may have to use the discrete phase model with UDF to solve it. hope could throw some light on the problem

 All times are GMT -4. The time now is 10:49.