A model with GaN still shows "No Active semiconductor materials were found"

device

#1

Hello KX Community,

I’m trying to simulate the charge transport of a 3D nanowire FinFETs with GaN, as indicated here:

I filled the upper part with Oxide and add 10V at Drain. The model can be meshed (although still getting a “mesh valid but may not be adequately refined”), but it shows “No active semiconductor materials were found” also. I used the doped GaN at the green part and GaN at the yellow part. The only problem that I can think of is that in the picure above the source and drain are the two sides of the same wire. In my model I used two little pads that are contact to the wire to simulate that.

My file is attached and I appreciate any form of support.
test_3D_nanowire_model_2.ldev (6.0 MB)

Thanks,
Yue


#2

You are right. The problem is arising from the fact that the source and drain contacts are getting merged with the “middle” GaAs region. When the solver creates the geometry by combining the different structures together, it combines similar materials that are connected to each other and creates a larger entity. In your setup, the “middle”, “source”, and “drain” rectangles are all getting connected together to form a large geometry for GaAs. Then when you are applying a boundary condition to either source or drain, the boundary condition gets applied to the entire GaAs region and since this is now a boundary condition, it is no longer part of the active region. This is the reason behind the error message.

The easiest way to resolve this is to use different materials for your source and drain. Since these will have boundary conditions applied, they will not be solver for transport any ways and so the material you assign to them is not important. Let me know if the problem goes away once you change the material of source and drain to something else than GaAs.