[drmazi]

Hi, i am modelling the formation of foam from vertically falling water jet containing dissolved surfactant falling unto a water surface resulting in formation of foam. I have been able to simulate falling water jet in Fluent but I do not know how to go about simulating the formation of foam bubbles. I need your suggestion and input.

[arjun]

it is little difficult to advise without really knowing how the foam is formed.



If the foam is simply a liquid with air trapped inside it then it is just shown by area where volume fraction is between 0 to 1.



Or foam is result of a chemical reaction resulting in a phase that has porosity then it is very tough to do with fluent. This video is where foam is generated due to chemical reaction:



https://youtu.be/9kkBMeEl2R8

[BlnPhoenix]

Just my to cents, open for correction and discussion:

To my knowledge this has not been done so far in realistic way using a VOF approach. The hindered Colescence is hard to implement, since on a microscopic level coalescence is only desribed accurately when you consider all the (inter-bubble-)forces acting on the bubbles that make up the foam (DLVO theory, zeta potential etc)...VOF usally tends to overpredict coalescence rates because these forces are not accounted for, so that in a VOF simulation no Foam will ever build up.
What you can work around is something like Arjun mendtioned, a third phase with certain properties that represents your foam on a macroscopic level in an Euler-Euler-type of simulation..

[drmazi]

Originally Posted by arjun

it is little difficult to advise without really knowing how the foam is formed.



If the foam is simply a liquid with air trapped inside it then it is just shown by area where volume fraction is between 0 to 1.



Or foam is result of a chemical reaction resulting in a phase that has porosity then it is very tough to do with fluent. This video is where foam is generated due to chemical reaction:



https://youtu.be/9kkBMeEl2R8

Yes, I know it a little difficult but the foam is formed as a result of air entrainment of the falling liquid which contains dissolved liquid soap.

There is no reaction involved. The foam is formed on the surface of the free water. I have conducted experiment to understand the physics and in my fluent simulation, I could only model bubble formation using Multiphase VOF.

[drmazi]

Originally Posted by BlnPhoenix

Just my to cents, open for correction and discussion:

To my knowledge this has not been done so far in realistic way using a VOF approach. The hindered Colescence is hard to implement, since on a microscopic level coalescence is only desribed accurately when you consider all the (inter-bubble-)forces acting on the bubbles that make up the foam (DLVO theory, zeta potential etc)...VOF usally tends to overpredict coalescence rates because these forces are not accounted for, so that in a VOF simulation no Foam will ever build up.
What you can work around is something like Arjun mendtioned, a third phase with certain properties that represents your foam on a macroscopic level in an Euler-Euler-type of simulation..

You are right. It has not been done. I am thinking if I could create a third phase specifying the liquid soap. But another consideration is that the falling jet falls onto a rectangular vessel of dimension 50cm x 30cm x 80 cm (length X width X height). The rectangular tank contains free water of height 10cm while the falling jet contains dissolved liquid soap.

Just seeking suggestions and advice. I have simulated using Multiphase VOF patching after initialization to account for the free water level.

My concern is how to specify the falling water jet to represent the dissolved liquid soap or incorporate it using UDF of empirical relation of foam height with time obtained from my experiment.

[BlnPhoenix]

Originally Posted by drmazi

You are right. It has not been done. I am thinking if I could create a third phase specifying the liquid soap. But another consideration is that the falling jet falls onto a rectangular vessel of dimension 50cm x 30cm x 80 cm (length X width X height). The rectangular tank contains free water of height 10cm while the falling jet contains dissolved liquid soap.

Just seeking suggestions and advice. I have simulated using Multiphase VOF patching after initialization to account for the free water level.

My concern is how to specify the falling water jet to represent the dissolved liquid soap or incorporate it using UDF of empirical relation of foam height with time obtained from my experiment.

About your concern: I would create an inlet on top of the tank, that releases the jet towards the surface. You can use a time depended inlet value for velocity to constrain the jet entering to a fixed time period.

[drmazi]

Originally Posted by BlnPhoenix

About your concern: I would create an inlet on top of the tank, that releases the jet towards the surface. You can use a time depended inlet value for velocity to constrain the jet entering to a fixed time period.

From my inquest, I think it is not possible to obtain chemical formula for Sodium Lauryl Sulfate in Fluent.

My thought is how to possibly include the experiment foam height variation with time using udf.

[BlnPhoenix]

Originally Posted by drmazi

From my inquest, I think it is not possible to obtain chemical formula for Sodium Lauryl Sulfate in Fluent.

My thought is how to possibly include the experiment foam height variation with time using udf.

With classical VOF approach you will not suceed imo. You will need to implement a pseudo force for each bubble acting as hinderance for bubble coalescence resulting in foam formation. Good luck

[arjun]

The way this simulation could be done is that one can run with 2 phase VOF simulation while third phase (foam) is generated as some model (or based on condition). One can solve simple scalar transport for this third phase with its generation controlled by VOF model.

[drmazi]

Originally Posted by arjun

The way this simulation could be done is that one can run with 2 phase VOF simulation while third phase (foam) is generated as some model (or based on condition). One can solve simple scalar transport for this third phase with its generation controlled by VOF model.

If h = 0.0048ln(t) + 0.0107 represent the empirical model for the foam height, please how do i implement and solve for third phase (foam). I am new to handling this.
Thanks

[BlnPhoenix]

Originally Posted by drmazi

If h = 0.0048ln(t) + 0.0107 represent the empirical model for the foam height, please how do i implement and solve for third phase (foam). I am new to handling this.
Thanks

You can create a mass source term for concentration of the third phase.

[arjun]

Originally Posted by drmazi

If h = 0.0048ln(t) + 0.0107 represent the empirical model for the foam height, please how do i implement and solve for third phase (foam). I am new to handling this.
Thanks

This is not really helpful.

What you are saying is that height is function of time. If this is exactly what you want then you can probably just overwrite the volume fraction to 1 based on this function. It gives you what you want for this 1 set up.

[drmazi]

Originally Posted by arjun

This is not really helpful.

What you are saying is that height is function of time. If this is exactly what you want then you can probably just overwrite the volume fraction to 1 based on this function. It gives you what you want for this 1 set up.

What I mean is the height of free water is fixed but as the water jet falls onto the free surface due to reduction in surface tension and air entrainment, foam gradually forms and the height of foam varies with time.

[BlnPhoenix]

Originally Posted by drmazi

What I mean is the height of free water is fixed but as the water jet falls onto the free surface due to reduction in surface tension and air entrainment, foam gradually forms and the height of foam varies with time.

The Problem with your equation, which should be fairly easy to implement into a mass source term, is that you only create foam and it will not be destroyed with time. In reality foam will eventually vanish, will it not?

[drmazi]

Originally Posted by BlnPhoenix

The Problem with your equation, which should be fairly easy to implement into a mass source term, is that you only create foam and it will not be destroyed with time. In reality foam will eventually vanish, will it not?

it takes very long time to varnish. At first creating foam is essential but very dense foam takes long time to varnish.

[BlnPhoenix]

Originally Posted by drmazi

it takes very long time to varnish. At first creating foam is essential but very dense foam takes long time to varnish.

Ok. Than to accomplish this write a Mass/Volume Fraction source term that creates an amount of volume of that "foam" phase according to your equation. Heigth of foam phase translates to a total amount of volume depending on your container geometry. This is straightfoward and easy to implement.
You may need an additional "sink term" to account for reduced volume of your other two phases during "foam" build up.

[Vasa]

does the viscosity increase? Can you provide a tutorial for making it? in my results the viscosity decreases below the viscosity of water, I really need information for my final college assignment

[JBeilke]

There is a nice work of Petr Karnakov and Sergey Litvinov on this topic. They also released their solver :-)

https://github.com/cselab/aphros
https://www.youtube.com/watch?v=iGdphpztCJQ
https://www.youtube.com/watch?v=0Cj8pPYNJGY

[Vasa]

Originally Posted by JBeilke

There is a nice work of Petr Karnakov and Sergey Litvinov on this topic. They also released their solver :-)

https://github.com/cselab/aphros
https://www.youtube.com/watch?v=iGdphpztCJQ
https://www.youtube.com/watch?v=0Cj8pPYNJGY

I'm still confused, whether in the Ansys fluent simulation foam formation with the VOF model can increase viscosity?