Verification of this phase change modeling
 

Hi everyone.

I'm trying to two phase flow with phase change using VOF method.
There are some approach for mass transfer modeling.
I'm interested in the following way.
mDot = K/L * grad(T)・grad(F)
Note that K refers to the mixture thermal conductivity, F refers to the volume fraction and L is the latent heat.

But I couldn't find that verification.
If you understand this correctry, please let me know the verification!
Thanks a lot.

Hi, Kawamura-san,

I can not make a verification for this notation.
But I think grad(F) should be grad(F)/mag(grad(F)).
This means unit normal vector of interface.

Hi Ohbuchi-san,

thanks for your response.
but FLUENT and some people show this expression in their paper .
there is non-existence of normal vector.

I think the most accurate expression is...
mDot = 1/L *(qg-ql)・n *Aint/V (kg/m^3 /s) in evaporating
where qg and ql is heat flux in gas and liquid phase respectively,Aint is interface area,V is volume of cell and n is unit normal vector.

but getting the correct interface area is so difficult in VOF.
so I think that grad(F) may works as n・Aint/V .
but i don't get a correct understanding about this :(

Hi. Kawamura-san,
The value of grad(F) will be very large on sharp interface, so I think some limiter function will be needed to restrict the value within 0-1.
And, I think it's impossible to determine interface area from grad(F) without assuming specific flow regime(such as stratified flow or bubbly flow).


BTW, the expression of mas transfer by heat transfer;
mDot = (k/L) grad(T) grad(F)/mag(grad(F)),
was originally introduced in following paper.

Son,G. and Dhir,V.K. A level set method for analysis of film boiling on an immersed solid surface. Nummer. Heat Transfer B,vol.52,pp.153-177.(2007)

This expression is only valid on stratified flow regime.

Hi Ohbuchi-san,

It's useful information for me!
Now it's starting to make sense.

I consider about assuming specific flow.

Thank you for your kindness.

Kawamura