I think you have to specify all the properties that change over time.density, specific heat capacity ,.......

Sepideh Bazari

What specification do you think is necessary for secondary phase viscosity when the whole mixture is modeled by non-Newtonian power law? (where viscosity depends on shear rate)

Are you solving for solid volume fraction in fluid? Is this your research objective? If so, Eulerian-Eulerian would be more convenient to work with. You could also use Discrete Phase Model (DPM) depending on your objectives. Since your secondary phase is solid, you will need to input the rheological properties of the primary fluid only: Consistency index, power law index, minimum and maximum viscosity. I hope this helps. Please let me know if you need any clarifications.

Dear Sinthuran Jegatheeswaran

No, the volume fractions are known.

Actually I'm considering a homogenous single-fluid mixture (assuming there is no slip velocity) and the purpose is to find out about how the whole mixture's non-Newtonian behaviour affects its hydrodynamic characteristics.

Since Fluent calculates the total density, thermal conductivity, viscosity, etc proportional to each phase's value multiplied by its volume fraction, I am wondering how to force Fluent to compute total viscosity only based on primary phase quantities given to it (non-Newtonian power law relation) but other properties in the usual way.

Should I put zero for secondary phase viscosity?

Your help would be greatly appreciated.

Since the secondary phase is solid, it's viscosity is zero.

I must mention that there are numerical limitations associated to each model.

If you consider a single-fluid mixture, the CFD model will solve for continuity and momentum equation for the single phase mixture. But, in reality, the hydrodynamics of solid-liquid mixture is more complicated and it requires more sophisticated model to capture its actual phenomena. In my opinion, a two-phase Eulerian-Eulerian model will be more accurate than mixture model.

How will you validate your numerical results? What microscopic validations are you considering?

Hi Ataollah Firoozi ,

Sorry for the delayed reply.

I believe that you are using a non-granular model. When you use a non-granular model, the slip velocity between the particles and the basefluid is not taken into account. Hence, the model will take into account only the density. All other properties which you may enter will be ignored.

But, based on my experience I wish to say that the results of a non-granular model will be almost same as that of a single-phase model.

As I see that you are modelling nanofluid flow, I suggest that using the granular model will be more suitable.

Please see this thread, where I have given some explanation on the definition of granular properties.

https://www.researchgate.net/post/How_to_specify_material_properties_for_Eulerian-Eulerian_multiphase_model_in_fluent

Thank You.

Sinthuran Jegatheeswaran

Well, in order to reduce computational cost and time I chose mixture model but I will try Eulerian-Eulerian too.

I was hoping to find an article in the literature somewhat similiar to my work and perform validations based on their results.

Thank you for your valuable comments.

Deepak Selvakumar R

I will try granular modelling as you indicated.

Thank you for your helpful comments.

Hi Ataollah Firoozi

If it is a batch process equipped with an impeller, try compare the experimentally measured torque with the numerically calculated value. If it is a continuous flow process, compare the pressure drop data. These are some of the coarse validations which gives a rough idea if the numerical results are reasonable and logical. It is always to good to have few microscopic validations such as velocity measurements, concentration profile etc to prove if the CFD captures the hydrodynamics accurate enough. The attached articles give some insights on the coarse and microscopic validations which can be employed to check the accuracy of the numerical predictions in batch and continuous processes.

I hope this helps.

Good luck with your research,

Sinthuran

Hi Ataollah Firoozi ,

You are welcome.

Regarding the non-Newtonian part of your question, I do not have any experience with the non-Newtonian models.

In my works, I have considered my nanofluids to be Newtonian fluids. This assumption is valid in a volume fraction