Full Multicomponent Diffusion

Diger · February 23, 2016, 06:47

So apparently a Full Multicomponent Diffusion is only possible for 3 or more species?! Why can't I simulate diffusion with 2 components?

CeesH · February 23, 2016, 06:53

Well, it's called multicomponent diffusion, right? So, there will have to be at least 2 diffusing species in a carrier; having 2 components, there is 1 diffusing species in a carrier.

Diger · February 23, 2016, 07:34

Yes I get that, but is diffusion then implicitly accounted for (maybe in fickian diffusion?) or not calculated at all? What alternatives do I have?
Though I don't understand why maxwell stefan equation is only valid for N>2 ? Shouldn't it go over to the simple 2 component fickian diffusion?

CeesH · February 23, 2016, 08:08

fickian diffusion is included in the species tracking model by default. Of course, you do need to give in a diffusion coefficient (under materials)

Diger · February 23, 2016, 08:59

Are you talking about the panel:
Species Model -> edit mixture material -> mass diffusivity (m2/s) ??

I specified kinetic theory so far...Though I do not quite get how ansys gets the right coefficients from that?!
manually would be which option:
constant-dilute-appx
dilute-approx
unity-lewis-number
multicomponent

?

CeesH · February 23, 2016, 09:53

I typically just throw in an experimental value, so constant dilute approx.; can't help you on the other options... sorry, hope someone else can.

LuckyTran · February 24, 2016, 02:19

You don't need to activate the multicomponent diffusion option if you only want to do diffusion of 2 species.

Diffusion with 2 components is generally not "called" multicomponent diffusion, it's diffusion.

February 23, 2016, 06:47	Full Multicomponent Diffusion	#1
Diger Senior Member Join Date: Nov 2015 Posts: 135 Rep Power: 11	So apparently a Full Multicomponent Diffusion is only possible for 3 or more species?! Why can't I simulate diffusion with 2 components?

February 24, 2016, 02:19		#7
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,752 Rep Power: 66	You don't need to activate the multicomponent diffusion option if you only want to do diffusion of 2 species. Diffusion with 2 components is generally not "called" multicomponent diffusion, it's diffusion. ehsan120 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Full Multi Component Diffusion is inactive	Alek	FLUENT	1	September 26, 2014 10:30
Moving mesh	Niklas Wikstrom (Wikstrom)	OpenFOAM Running, Solving & CFD	122	June 15, 2014 07:20
Multicomponent diffusion modeling	gsbhatia89	CFX	2	June 8, 2010 19:31
multicomponent diffusion	katya	FLUENT	0	November 24, 2004 11:31

February 23, 2016, 06:53		#2
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	Well, it's called multicomponent diffusion, right? So, there will have to be at least 2 diffusing species in a carrier; having 2 components, there is 1 diffusing species in a carrier.

February 23, 2016, 07:34		#3
Diger Senior Member Join Date: Nov 2015 Posts: 135 Rep Power: 11	Yes I get that, but is diffusion then implicitly accounted for (maybe in fickian diffusion?) or not calculated at all? What alternatives do I have? Though I don't understand why maxwell stefan equation is only valid for N>2 ? Shouldn't it go over to the simple 2 component fickian diffusion?

February 23, 2016, 08:08		#4
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	fickian diffusion is included in the species tracking model by default. Of course, you do need to give in a diffusion coefficient (under materials)

February 23, 2016, 08:59		#5
Diger Senior Member Join Date: Nov 2015 Posts: 135 Rep Power: 11	Are you talking about the panel: Species Model -> edit mixture material -> mass diffusivity (m2/s) ?? I specified kinetic theory so far...Though I do not quite get how ansys gets the right coefficients from that?! manually would be which option: constant-dilute-appx dilute-approx unity-lewis-number multicomponent ?

February 23, 2016, 09:53		#6
CeesH Senior Member Cees Haringa Join Date: May 2013 Location: Delft Posts: 607 Rep Power: 0	I typically just throw in an experimental value, so constant dilute approx.; can't help you on the other options... sorry, hope someone else can.