Material diffusion coefficient in OpenFOAM

Kikuta.K · July 13, 2018, 05:27

Hello Foamers.

I'm trying to add selective diffusion of hydrogen to reactingFoam, but I don't know where I can edit material diffusion coefficient.
Is diffusion coefficient calculated from Sutherland coefficient?
How can I edit and add selective diffusion?

I am sorry for not being excellent in OpenFOAM yet.

Hopefully anyone could help me to fix this problem?
My OpenFOAM is version 4.x.

Best regards.

clapointe · July 13, 2018, 15:28

reactingFoam uses the effective viscosity for species diffusion -- if you use sutherland viscosity then yes, it'd be used for computing (bulk) species diffusivity in the laplacian term. If you want to add differential diffusion, you'll need to compute diffusion coefficients for each species. There are various threads on this -- here's one for example : Difficulties with premixed laminar methane flame in openFoam.

Caelan

Kikuta.K · July 16, 2018, 08:43

Hi clapointe.

Thank you for your quick reply. (and I'm sorry for my late reply)

I have read the thread which you indicated.
Thank you for providing me another thread.

I've tried to imitate the code which was indicated by Chris to be able to calculate species diffusion, but didn't work well because of the difference of the solver.

Do you know another similar thread in reactingFoam?
If it's possible, I want you to provide me any other informations.

Thank you!

clapointe · July 16, 2018, 12:30

There really is not much difference between reactingFoam and fireFoam if fireFoam is used solely as a solver for reacting flow -- assuming a turbulent Prandtl number of 1 will pretty much make muEff = alphaEff, and the hEqn's are identical. This still assumed a unity Lewis No -- you can compute a bulk species diffusivity for non-unity Lewis No as done by FM global in their fireFoam repository : https://github.com/fireFoam-dev/fire...solver/YEEqn.H. Computing species-specific diffusion coefficients will be more involved -- see eg. : https://www3.nd.edu/~powers/ame.60636/transport.pdf.

In the thread I referenced, it looks like the Lewis No for each species is being taken into account.

Caelan

Gaozw · July 29, 2018, 22:10

Hi, foamers,

I am also making efforts to add speice diffusivity models to reactingFoam. Another thread on this is Multi species mass transport library [update] . The code runs well. However, I have no idea how to validate it.

clapointe · July 29, 2018, 22:20

One option could be finding DNS results (eg flame speed data) for which they compute species diffusivity and see how your implementation compares.

Caelan

Kikuta.K · August 2, 2018, 21:00

Quote:

Originally Posted by Gaozw

Hi, foamers,

I am also making efforts to add speice diffusivity models to reactingFoam. Another thread on this is Multi species mass transport library [update] . The code runs well. However, I have no idea how to validate it.

Hi Gaozw

I have already tried to compile "modifiedReactingFoam" which is indicated in the URL you posted, but couldn't.
is it due to the difference in version of OpenFOAM?
My OpenFOAM is version 4.1.

Which version are you using?
Also, did you compile modifiedReactingFoam without editing anything?

Regards,

Kikuta

Gaozw · August 7, 2018, 00:32

Hi Kikuta,
I am using OpenFOAM-2.3.1, which matches the author's latest release version. I follow the installation guide and it works fine. It may be incompatible with version-4.1 since there are many differeces between the two versions.

Gaozw · August 7, 2018, 01:25

Hi clapointe,

Comparing with DNS results is always a good choice. However, it is difficult for me because the diffusivity data in the combustion process can be barely found. I have to compromise and compare the combustion characteristic properties (i.e. flame thickness, flame speed, etc.). Is there a better way?

PS--sorry for the late reply.

July 13, 2018, 05:27	Material diffusion coefficient in OpenFOAM	#1
Kikuta.K New Member Kikuta Enoki Join Date: Jun 2017 Posts: 9 Rep Power: 9	Hello Foamers. I'm trying to add selective diffusion of hydrogen to reactingFoam, but I don't know where I can edit material diffusion coefficient. Is diffusion coefficient calculated from Sutherland coefficient? How can I edit and add selective diffusion? I am sorry for not being excellent in OpenFOAM yet. Hopefully anyone could help me to fix this problem? My OpenFOAM is version 4.x. Best regards.

July 16, 2018, 12:30		#4
clapointe Senior Member Join Date: Aug 2015 Posts: 494 Rep Power: 15	There really is not much difference between reactingFoam and fireFoam if fireFoam is used solely as a solver for reacting flow -- assuming a turbulent Prandtl number of 1 will pretty much make muEff = alphaEff, and the hEqn's are identical. This still assumed a unity Lewis No -- you can compute a bulk species diffusivity for non-unity Lewis No as done by FM global in their fireFoam repository : https://github.com/fireFoam-dev/fire...solver/YEEqn.H. Computing species-specific diffusion coefficients will be more involved -- see eg. : https://www3.nd.edu/~powers/ame.60636/transport.pdf. In the thread I referenced, it looks like the Lewis No for each species is being taken into account. Caelan Gaozw likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Frequently Asked Questions about Installing OpenFOAM	wyldckat	OpenFOAM Installation	3	November 14, 2023 12:58
Adding material to CFX material library	marlon	CFX	2	October 16, 2016 10:33
Compression stoke is giving higher pressure than calculated	nickjuana	CFX	62	May 19, 2015 14:32
diffusion coefficient	mamalili	CFX	8	February 24, 2015 17:51
question about how to define diffusion coefficient in CFX	rystokes	CFX	0	December 12, 2009 06:32

July 13, 2018, 15:28		#2
clapointe Senior Member Join Date: Aug 2015 Posts: 494 Rep Power: 15	reactingFoam uses the effective viscosity for species diffusion -- if you use sutherland viscosity then yes, it'd be used for computing (bulk) species diffusivity in the laplacian term. If you want to add differential diffusion, you'll need to compute diffusion coefficients for each species. There are various threads on this -- here's one for example : Difficulties with premixed laminar methane flame in openFoam. Caelan

July 16, 2018, 08:43		#3
Kikuta.K New Member Kikuta Enoki Join Date: Jun 2017 Posts: 9 Rep Power: 9	Hi clapointe. Thank you for your quick reply. (and I'm sorry for my late reply) I have read the thread which you indicated. Thank you for providing me another thread. I've tried to imitate the code which was indicated by Chris to be able to calculate species diffusion, but didn't work well because of the difference of the solver. Do you know another similar thread in reactingFoam? If it's possible, I want you to provide me any other informations. Thank you!

July 29, 2018, 22:10		#5
Gaozw New Member Gao Zhengwei Join Date: Jan 2017 Location: HangZhou, P.R.China Posts: 10 Rep Power: 9	Hi, foamers, I am also making efforts to add speice diffusivity models to reactingFoam. Another thread on this is Multi species mass transport library [update] . The code runs well. However, I have no idea how to validate it.

July 29, 2018, 22:20		#6
clapointe Senior Member Join Date: Aug 2015 Posts: 494 Rep Power: 15	One option could be finding DNS results (eg flame speed data) for which they compute species diffusivity and see how your implementation compares. Caelan

August 7, 2018, 00:32		#8
Gaozw New Member Gao Zhengwei Join Date: Jan 2017 Location: HangZhou, P.R.China Posts: 10 Rep Power: 9	Hi Kikuta, I am using OpenFOAM-2.3.1, which matches the author's latest release version. I follow the installation guide and it works fine. It may be incompatible with version-4.1 since there are many differeces between the two versions.

August 7, 2018, 01:25		#9
Gaozw New Member Gao Zhengwei Join Date: Jan 2017 Location: HangZhou, P.R.China Posts: 10 Rep Power: 9	Hi clapointe, Comparing with DNS results is always a good choice. However, it is difficult for me because the diffusivity data in the combustion process can be barely found. I have to compromise and compare the combustion characteristic properties (i.e. flame thickness, flame speed, etc.). Is there a better way? PS--sorry for the late reply.