Formation of oxide

cmageski · June 29, 2012, 10:41

Hello, gays, I wonder if I can simulate the formation of oxide in a tube flowing water saturated with oxygen, and how could you do this, since the oxide is a solid material.
If not.
How do I implement a thickness for the oxide, in the pipeline to set boundary conditions for diffusion of oxygen in the oxide that are different from the fluid.

ghorrocks · July 1, 2012, 08:32

How does the oxide form? Does oxygen react with gas/liquid and precipitate out or does it react with the vessel wall and form an oxide skin?

And what do you want to model with this chemistry? The surface roughening? Or the wall thinning? Or the fluid domain getting choked with oxide? Or just how thick the oxide becomes?

cmageski · July 2, 2012, 08:21

It react with the vessel wall and form an oxide skin.
I want the concentration oxigen on the surface this oxide and the thickness of it after a certain time. In order to predict the corrosion rate this pipe.

ghorrocks · July 2, 2012, 08:39

CFX can model all the basic physics and chemistry in this process, you just need to specify it. What information do you have about corrosion rate versus oxygen concentration, and any other parameters (eg corrosion rate versus depth of existing oxide)?

cmageski · July 2, 2012, 09:25

I have an equation that depends on the concentration of oxygen, and I need os concentration in the first node near the wall. although with the formation of oxide position of the node and concentration also changes. In susface of the oxide the velocity is zero, but the concentration not. What I export of the CFX is a concentration of oxygen carried by water through the additional variable in developed turbulent flow at the entrance. But to provide the formation of the oxide I'm not getting.

Please help me!

ghorrocks · July 2, 2012, 19:09

Do you have a model for how the existing oxide affects new oxide formation?

I suspect you will need an AV to track the oxide thickness. Ideally this would be an AV on the wall only but I cannot see how to implement that so it will probably have to be a field AV, but one which you only define values at the wall.

cmageski · July 3, 2012, 07:46

But you know how can I todo this? Do you have more information about this field AV?

cmageski · July 3, 2012, 07:53

Sorry I forgot to answer first question. I don't have a model, but of the oxide only is formed. It don't exist before and after of formed it don't affect new oxide formation.

ghorrocks · July 3, 2012, 18:59

The tutorials have some examples of AVs (I think).

Sounds like a simple case of integrating the oxygen concentration over time then. But unfortunately this is tricky to implement in CFX - you can either do it in fortran, or use a sneaky trick to allow you to access the value of an AV in the previous time step (search the forum for TRANS_LOOP and I think you will find it).

cmageski · July 24, 2012, 13:09

I have a problem with transport in turbulent flow of additional variable where the pressure grows and stabilizes above the speed which means that there is no convergence problem.
What type of solution is the outlet pressure of 1 atm.
The geometry is a pipe with contraction and expansion.
Please help me.

ghorrocks · July 24, 2012, 20:56

Can you post this as a new thread? No need to hijack other threads.

cmageski · July 25, 2012, 13:28

Yes I did this. When I realized I was posting here it was late.

June 29, 2012, 10:41	Formation of oxide	#1
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	Hello, gays, I wonder if I can simulate the formation of oxide in a tube flowing water saturated with oxygen, and how could you do this, since the oxide is a solid material. If not. How do I implement a thickness for the oxide, in the pipeline to set boundary conditions for diffusion of oxygen in the oxide that are different from the fluid.

July 24, 2012, 13:09	Problem in pressure	#10
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	I have a problem with transport in turbulent flow of additional variable where the pressure grows and stabilizes above the speed which means that there is no convergence problem. What type of solution is the outlet pressure of 1 atm. The geometry is a pipe with contraction and expansion. Please help me.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Ice Particle formation using CFX	Dinesh	CFX	1	April 7, 2010 17:39
FLuent Sulphur formation	jehuty	FLUENT	1	April 14, 2009 03:07
VOF and Droplet Formation	saang	Siemens	1	June 7, 2005 03:59
NOx Formation in Turbulent hydrogen combustion	Ali Yigit	FLUENT	1	December 22, 2003 03:50
SOx formation	olivier	Main CFD Forum	0	December 18, 2001 12:27

July 1, 2012, 08:32		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	How does the oxide form? Does oxygen react with gas/liquid and precipitate out or does it react with the vessel wall and form an oxide skin? And what do you want to model with this chemistry? The surface roughening? Or the wall thinning? Or the fluid domain getting choked with oxide? Or just how thick the oxide becomes?

July 2, 2012, 08:21		#3
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	It react with the vessel wall and form an oxide skin. I want the concentration oxigen on the surface this oxide and the thickness of it after a certain time. In order to predict the corrosion rate this pipe.

July 2, 2012, 08:39		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	CFX can model all the basic physics and chemistry in this process, you just need to specify it. What information do you have about corrosion rate versus oxygen concentration, and any other parameters (eg corrosion rate versus depth of existing oxide)?

July 2, 2012, 09:25		#5
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	I have an equation that depends on the concentration of oxygen, and I need os concentration in the first node near the wall. although with the formation of oxide position of the node and concentration also changes. In susface of the oxide the velocity is zero, but the concentration not. What I export of the CFX is a concentration of oxygen carried by water through the additional variable in developed turbulent flow at the entrance. But to provide the formation of the oxide I'm not getting. Please help me!

July 2, 2012, 19:09		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Do you have a model for how the existing oxide affects new oxide formation? I suspect you will need an AV to track the oxide thickness. Ideally this would be an AV on the wall only but I cannot see how to implement that so it will probably have to be a field AV, but one which you only define values at the wall.

July 3, 2012, 07:46		#7
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	But you know how can I todo this? Do you have more information about this field AV?

July 3, 2012, 07:53		#8
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	Sorry I forgot to answer first question. I don't have a model, but of the oxide only is formed. It don't exist before and after of formed it don't affect new oxide formation.

July 3, 2012, 18:59		#9
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	The tutorials have some examples of AVs (I think). Sounds like a simple case of integrating the oxygen concentration over time then. But unfortunately this is tricky to implement in CFX - you can either do it in fortran, or use a sneaky trick to allow you to access the value of an AV in the previous time step (search the forum for TRANS_LOOP and I think you will find it).

July 24, 2012, 20:56		#11
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Can you post this as a new thread? No need to hijack other threads.

July 25, 2012, 13:28		#12
cmageski New Member Cecília Mageski Join Date: Sep 2011 Posts: 16 Rep Power: 14	Yes I did this. When I realized I was posting here it was late.