[rodrigovime]

I need to assign the thermal conductivity of the "layer" of cells next to the boundary to the adjacent "layer" of cells inside the domain.
In the attached image, it will be something like assign the thermal conductivity of the orange line to the green line.
Does anyone know how to do it or if is it possible?
Thanks for your response.

[Gert-Jan]

If it are different solids, then you should create a volume with only the orange elements, assign it to a new material with a different conductivity.....If you have a liquid and are trying to modify the conductivity close to the wall, then there must be a specific reason. If so, then what are you trying to do exactly?

[rodrigovime]

I'm trying to decrease an enormous gradient near the wall by making thermal conductivity artificially high. The huge gradient of temperature is created by resistive heating due to an electrical current that passes through the fluid. I've tried a constant value, but it doesn't work properly.

[evcelica]

It does not sound like what you are trying to do is realistic. The artificially high conductivity of the fluid is not realistic, and it any results obtained using it would be meaningless, so i don't understand the reasoning for trying to do this.
But If you really wanted to, you could probably make the conductivity a function of y (or another location based definition)

[rodrigovime]

Thanks for your reply.
Yes, it's not realistic, but I'm trying to see if it works. I'll try the location based approach.

[ghorrocks]

This sounds like a classic XY problem to me (https://en.wikipedia.org/wiki/XY_problem)

Why do you think this gradient is unrealistically high? Is it is problem anyway? If you are just adding heat to a liquid it might not matter if the gradients are huge.

Should there be other physics at play (eg boiling?). Do you need a finer mesh to resolve it properly? Or maybe a turbulence model?