Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Evaporation/Dissolution of a sphere and moving mesh (ALE): results not matching theory

Etienne Jambon-Puillet

Please login with a confirmed email address before reporting spam

Hi I want to simulate the dissolution/evaporation of various shaped objects. As a first test I did the simplest one I know which is solvable analytically: the quasi-static evaporation of a sphere by pure diffusion.

To do so I use the transport of diluted species (tds) module with a moving mesh (ale) in a time dependent study. I generate a big sphere with a spherical hole inside, then calculate the concentration field with the tds module (convection box unchecked and stationary equations, concentration set to some value at the hole surface and zero on the outer sphere ) and prescribe a normal velocity to the surface of the hole equal to the normal gradient of the concentration.

At each time step the concentration field is correct (decay as 1/r) but I have problems with the moving mesh:
- The mesh moves in 2D or 3D but comsol fails to converge (reach singularity) in 2D axisymmetric which is weird and force me to go 3D. Why, is there a way out ?
- After several steps the hole is not spherical anymore, I guess it is to be expected due to the sum of small errors. Is there a way to do better except refining the mesh (impossible in 3D with my computer) ?
- The rate of shrinkage of the hole is much slower than the analytical prediction.

Any suggestion regarding these problems ?
Thank you.


1 Reply Last Post 22 giu 2017, 05:04 GMT-4
Etienne Jambon-Puillet

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 22 giu 2017, 05:04 GMT-4
Updated: 7 years ago 22 giu 2017, 09:19 GMT-4
Ok so I solved my issue with the 2D axisymmetric version. The problem was the boundary at r=0 connected to the hole which moves. Allowing this boundary to extend in the z direction while the hole shrinks solved it.

And for the match with the analytical solution, it improves while improving the mesh so I guess I know what to do.
Ok so I solved my issue with the 2D axisymmetric version. The problem was the boundary at r=0 connected to the hole which moves. Allowing this boundary to extend in the z direction while the hole shrinks solved it. And for the match with the analytical solution, it improves while improving the mesh so I guess I know what to do.

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.