Learning Forum

[gloria.nulchis]

Dear Ansys community,

I am wondering whether blood is registered as a material in a data base for CFD? I am aware that data for viscosity and shear rate is available. But this is just incomplete mimicking of blood properties instead of creating a material that simulates the real world conditions: Millions of small viscoelastic particels (erythrocytes, other blood cells can be neglected because of numbers) embedded in a newtonian fluid (plasma) with special flow characteristics.

That would be important, because blood behaves oddly in vessels which diameters are between 5 and 300 micrometers.  In this range of diameters the apparent viscosity of blood is decreasing with the diameter of the vessel. Thats because of the Fahraus-Lindqvist-Effekt: In vessels with small diameters cells align perfectly in the middle of the vessels with plasma surrounding it, resulting in a decreased viscosity. When the diameters of the vessels dip below the diameter of the blood cells (about 7 mikrometers) the cells deform and slip through nevertheless. Only when the vessel diameter gets as narrow as 3-4 micrometers erythrocites can no longer pass which leads to a sudden increase in apparent viscosity. I have not found a material in any database that resembles these quite unique features of blood properly.

Has anybody ever created a material like that? And if not, is it possible to create a material like blood for CFD?

Any help would be much appreciated, thank you very much in advance!

[Rob]

Treading carefully to avoid breaking any of the guidelines. 

Blood can be modelled at a bulk level using a nonNewtonian viscosity. There are a few options, but the one I tend to use is based on  Gijsen F.J.H., van de Vosse F.N., Janssen J.D.; 1999;
The Influence of the non-Newtonian properties of blood on the flow in large arteries: steady flow in a carotid bifurcation model; Jou of Biomechanics, Vol 32, pp601     If your supervisor has access to the Ansys Portal the UDF should be available as a solution: I cannot share here. 

However, at very small scale (as you note) blood is plama and lots of floating bits. So, I assume at below 300microns it behaves as a carrier fluid with bits in it. How big are red cells etc? 

[gloria.nulchis]

 

Hi Rob, thank you very much for your reply.

What is an UDF in this context? Red blood cells are biconcave elastic discs with diameters of appoximately 7,5 microns. Other blood cells are rather rare in comparison to red blood cells, therefore their contribution to the viscoelastic nature of blood is negtectable. One microliter blood contains about 5 million red blood cells (6×10^12cells/liter = 6 trillion cells/liter). Is it possible to create a fluid like that for simulation purposes?

 

[Rob]

User Defined Function: the blood viscosity model wasn't in the code so I wrote it as a user function to do it. That's published as a solution, but only accessable to the licenced users: so not Student users or students (software and people respectively). 

With a channel diameter of 5-300 microns the red cells go from squeezing through to lubricated blockages to free stream motion so I can see why the "effective viscosity" would change. It's not so much a viscosity change as a more complex change in flow phenomena. Fluent doesn't have a model for that, but it could be written with some caveats.