CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   OpenFoam - Fixed Bed Reactor (https://www.cfd-online.com/Forums/openfoam-solving/255364-openfoam-fixed-bed-reactor.html)

InesR April 4, 2024 06:12

OpenFoam - Fixed Bed Reactor
 
Hello! I've been learning recently the OpenFoam software for my Master's Degree in Chemical Engineering. My problem consists of studying the hydrodynamics of a fixed bed reactor , with catalysts particles of different geometries, and I've encounter some doubts and wanted an opinion:

1- I've learned that we cancreate fixed bed reactors using other softwares (e.g. Blender) and export as a single .stl file to OpenFoam, however, isn't better to create two separate .stl files (one for reactor geometry and another one for the catalyst) and export to OpenFoam?

2- Still about point 1, I've been trying to search for tutorials but haven't gotten any luck, with OpenFoam is possible to determine how many particles I want and simulate them going into the reactor?
I already created the catalysts .stl file (1 catalyst of a certain geometry, needed to define a number for simulation)

3- I've seen in this forum some people with the same problem as me and wanted to know if Eulerian approach was the most suitable for this type of problem (particles in the same position in the beggining and end of simulation). Another approach I've been seeing alot if twoPhaseEulerFoam, which one better?
If there is any type of different approach that you suggest I'd appreciate!

4- For the particles, beside Eulerian approach, probably better assuming the catalysts as porous media?

Thank you all for the availability, and sorry in advance if I've made some questions that may seem confusing (still trying to learn the software and what type of approach I can use for my study case).

saidc. April 4, 2024 07:46

Quote:

Originally Posted by InesR (Post 867227)
Hello! I've been learning recently the OpenFoam software for my Master's Degree in Chemical Engineering. My problem consists of studying the hydrodynamics of a fixed bed reactor , with catalysts particles of different geometries, and I've encounter some doubts and wanted an opinion:

1- I've learned that we cancreate fixed bed reactors using other softwares (e.g. Blender) and export as a single .stl file to OpenFoam, however, isn't better to create two separate .stl files (one for reactor geometry and another one for the catalyst) and export to OpenFoam?

2- Still about point 1, I've been trying to search for tutorials but haven't gotten any luck, with OpenFoam is possible to determine how many particles I want and simulate them going into the reactor?
I already created the catalysts .stl file (1 catalyst of a certain geometry, needed to define a number for simulation)

3- I've seen in this forum some people with the same problem as me and wanted to know if Eulerian approach was the most suitable for this type of problem (particles in the same position in the beggining and end of simulation). Another approach I've been seeing alot if twoPhaseEulerFoam, which one better?
If there is any type of different approach that you suggest I'd appreciate!

4- For the particles, beside Eulerian approach, probably better assuming the catalysts as porous media?

Thank you all for the availability, and sorry in advance if I've made some questions that may seem confusing (still trying to learn the software and what type of approach I can use for my study case).

Hi Ines,

For the first problem you mentioned, you can benefit from DEM (Discrete Element Method) or Monte Carlo methods. You can create a packed bed dependent on porosity or the number of particles (spherical) with cemfDEM and OpenMC open-source codes (easy to use). You can solidify the coordinates you obtain using a simple block of code in SALOME TUI and then mesh it in 2D and export it as STL. Then, you can create the background (container) mesh with blockMesh, and finally obtain the mesh you want with snappyHexMesh.

However, in this method, you need to pay close attention to the contact points. Because creating a solvable mesh in those areas is a laborious task. There are quite a lot of studies on this subject.

OpenMC: Link
cemfDEM: Link

regards,
Said

InesR April 4, 2024 08:00

Hello. Thank you for the fast response, I'll do more study and try to see the DEM method, I've read it in some articles applied to packed bed reactors.


All times are GMT -4. The time now is 22:37.