Author Topic: Boundary Conditions for above and below the gate for my DEVICE?  (Read 9609 times)

0 Members and 1 Guest are viewing this topic.

Offline Ash

  • Regular QuantumATK user
  • **
  • Posts: 24
  • Country: in
  • Reputation: 0
    • View Profile
Can anyone tell me the general criteria for about the boundary conditions for device with gate enabled? I am using the Boundary condition as shown in the attached file corresponding to my device geometry. Is it correct ?
Can you please tell me the effect of dirichlet boundary conditions at the top of the gate (top(B)) and dirichlet at the bottom(bottom(B))?
« Last Edit: May 8, 2012, 12:37 by Ash »

Offline hzkvictory

  • Regular QuantumATK user
  • **
  • Posts: 23
  • Country: cn
  • Reputation: 0
    • View Profile
I am very interested in this problem as well.
according to  http://quantumwise.com/documents/tutorials/latest/GrapheneDevice/index.html/chap.transmission_spectrum.html#vnl.transmission_spectrum.save
It seems that the device all have Diriklet condition on C direction, and the metallic and dielectric region needs Neumann condition on A,B direction.

and  I would like to know if the device is constructed wiht graphene sheet (not graphene ribbon as above shows) and spatial region. what would it be?

Offline Nordland

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 812
  • Reputation: 18
    • View Profile
Can anyone tell me the general criteria for about the boundary conditions for device with gate enabled? I am using the Boundary condition as shown in the attached file corresponding to my device geometry. Is it correct ?
Can you please tell me the effect of dirichlet boundary conditions at the top of the gate (top(B)) and dirichlet at the bottom(bottom(B))?

Hey Ash.

I digged out a calculation I did some time ago of a similar system, that might spread some light on this matter. Starting with the bottom gate:
since your gate is pretty much covering the entire bottom face (making it effectively dirichlet), you will only see little effect of what you choose here. If you set it to dirichlet, then every point in the bottom that is not covered by your gate will be forced to zero. This will make the effect of the gate very localized in space in the A,B plane, but all in all this does not really matter in this simulation.

The choice for the top plane matters a lot. If you set it to dirichlet, you force the electrostatic potential to be zero at the top and by doing so, you will force the electric field strength to be gate voltage over length of cell in that direction. The electric field strength would be very strong, in other word the gate potentials has too drop off much faster. If you set it to Neumann, the field strength can vary and the potential drop off can be as long as required,
so if the desired is to model vacuum at the top, i would personally go with Neumann there.

Offline Ash

  • Regular QuantumATK user
  • **
  • Posts: 24
  • Country: in
  • Reputation: 0
    • View Profile
Thanks Norland, for detailed and effective reply !!  :)