Multiphase flow with two continuous fluids and one dispersed fluid

ashtonJ · July 24, 2014, 00:12

Dear all,
I am trying to model the mixing of foam with blood in varicose veins. The model consists of two parts; needle and vein as shown in the attached image. Foam and air bubbles with velocity of 1.88 [m/s] enters the vein which initially contains stationary blood. I set a zero pressure at all outlets of the vein to have a zero blood flow in the vein. What I want to do is to let foam and air bubbles flowing in the needle enter the vein which contain stationary blood. In setting up in CFX Pre, I created two domains; one for needle and one for vein. In basic setting > Fluid and Particle Definitions tab for needle, I defined two items; foam as a continuous fluid and air as a dispersed fluid. Then I realized that I have these items also in setting > Fluid and Particle Definitions tab for vein, however, in the vein I want to have only blood because vein is initially filled with blood.
Then, I added another item in setting > Fluid and Particle Definitions as blood. Therefore, in both needle and vein domains, three fluids were present; foam, blood and air. At the beginning of the solution I got the error that “each dispersed fluid can only have one continuous fluid”.

Please can anyone help me to set up CFX-Pre for this problem.
Thanks
AshtonJ

ghorrocks · July 24, 2014, 06:55

This sounds like a single domain simulation to me. Should be no need for multiple domains.

You need to be completely clear what you are modelling. Does the air enter as foam? Then just model blood and air - with air probably being a disperse phase. Note that disperse phase models have volume fraction limits so you will have to make sure your foam is not too aerated for your model to be valid.

If you are talking about a highly aerated foam then I am not aware of a multiphase model which can handle this. (Either in CFX or any other solver)

ashtonJ · July 24, 2014, 10:31

Dear Glenn,
Thank you.
Foam is a liquid containing air bubbles. The foam and air bubbles enter the vein through the needle with a constant velocity. The vein initially contains blood which is stationary. In my model, the needle should only include liquid foam and air bubbles and vein include blood. I want to see how far the bubbles travel in the vein.

ghorrocks · July 24, 2014, 19:32

Air mixed in a liquid has many different states. When the volume fraction is low the air bubbles float around largely independent of each other, so the motion of the bubbles is predominantly determined by the interphase drag. This can be modelled using the multiphase models available in CFX. At high volume fraction the air bubbles are pushed up against each other and the motion of the air becomes increasingly determined by the bubbles acting directly on each other. Other effects include the bubbles becoming non-spherical and the liquid between the bubbles thinning to just a film. At very high volume fractions we call this a foam and CFX does not have models to handle this.

At moderate volume fractions you get a bit of both effects. CFX might be able to model this using high volume fraction multiphase models.

These are the questions you should be looking into to determine if you can model this.

ashtonJ · July 25, 2014, 00:21

Thank you Glenn. As suggested, it is better to start with the medium volume fraction case but I still have two questions:
1- Needle only contain foam with bubble and vein contains stationary blood. How I can make the CFX understand that in needle there is foam with bubble and in vein there is only blood.
2- When I define two continous liquids (foam and blood) and a disperesed fluid (air bubbles), I get an error indicating that "each dispersed fluid can only have one continuous fluid".

ghorrocks · July 25, 2014, 07:31

Before you proceed, do you have any idea what low, medium and high volume fraction actually is? I mean the volume fractions which define these regimes? You have to look in the CFX documentation to determine what volume fraction ranges the available models are suitable for. If your volume fraction is going to be out of that range then you cannot model it in CFX.

Q1 - This is a basic question on multiphase flow. If you cannot answer this yourself then there is no way you know enough of what is going on to complete a simulation. You need to do more tutorials and reading to understand multiphase flow before proceeding.

Q2 - This is also a very basic question on multiphase flow - you really need to do some study on this topic. Blood will be the continuous phase, and air will be the disperse phase. There are possibly other approaches but this is most obvious one.

July 24, 2014, 06:55		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	This sounds like a single domain simulation to me. Should be no need for multiple domains. You need to be completely clear what you are modelling. Does the air enter as foam? Then just model blood and air - with air probably being a disperse phase. Note that disperse phase models have volume fraction limits so you will have to make sure your foam is not too aerated for your model to be valid. If you are talking about a highly aerated foam then I am not aware of a multiphase model which can handle this. (Either in CFX or any other solver)

July 24, 2014, 10:31		#3
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 17	Dear Glenn, Thank you. Foam is a liquid containing air bubbles. The foam and air bubbles enter the vein through the needle with a constant velocity. The vein initially contains blood which is stationary. In my model, the needle should only include liquid foam and air bubbles and vein include blood. I want to see how far the bubbles travel in the vein.

July 24, 2014, 19:32		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	Air mixed in a liquid has many different states. When the volume fraction is low the air bubbles float around largely independent of each other, so the motion of the bubbles is predominantly determined by the interphase drag. This can be modelled using the multiphase models available in CFX. At high volume fraction the air bubbles are pushed up against each other and the motion of the air becomes increasingly determined by the bubbles acting directly on each other. Other effects include the bubbles becoming non-spherical and the liquid between the bubbles thinning to just a film. At very high volume fractions we call this a foam and CFX does not have models to handle this. At moderate volume fractions you get a bit of both effects. CFX might be able to model this using high volume fraction multiphase models. These are the questions you should be looking into to determine if you can model this.

July 25, 2014, 00:21		#5
ashtonJ Senior Member Ashkan Javadzadegan Join Date: Sep 2010 Posts: 255 Rep Power: 17	Thank you Glenn. As suggested, it is better to start with the medium volume fraction case but I still have two questions: 1- Needle only contain foam with bubble and vein contains stationary blood. How I can make the CFX understand that in needle there is foam with bubble and in vein there is only blood. 2- When I define two continous liquids (foam and blood) and a disperesed fluid (air bubbles), I get an error indicating that "each dispersed fluid can only have one continuous fluid".

July 25, 2014, 07:31		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,854 Rep Power: 144	Before you proceed, do you have any idea what low, medium and high volume fraction actually is? I mean the volume fractions which define these regimes? You have to look in the CFX documentation to determine what volume fraction ranges the available models are suitable for. If your volume fraction is going to be out of that range then you cannot model it in CFX. Q1 - This is a basic question on multiphase flow. If you cannot answer this yourself then there is no way you know enough of what is going on to complete a simulation. You need to do more tutorials and reading to understand multiphase flow before proceeding. Q2 - This is also a very basic question on multiphase flow - you really need to do some study on this topic. Blood will be the continuous phase, and air will be the disperse phase. There are possibly other approaches but this is most obvious one.