Multicomponent diffusion coefficient

[Simon Lorentzen]

Dear all, 

I have some trouble regarding the use of multicomponent transport, and I was hoping some of you could help me.

Firstly, is it possible to extract the multicomponent diffusion coefficient with the spatial changes along a flame? As the temperature changes along the axial position, I would assume that the coefficient does as well? 

Secondly, every time i use solve it seems like the transport model reverts to mix. Is there some way I can avoid this? 

I have attached a small working example that shows that the transport model changes. 

Kindest regards, and thank you in advance,

Simon