Spatial Dielectric/ Metallic Region

[Dipankar Saha]

Hi,

1) Is it necessary to extend the spatial dielectric region into the electrodes too?

2) In the following tutorial, how do you get the del_VH as a solid curve (I mean, there has been observed no oscillatory  nature..which generally comes from the distrbn. of the atoms along the axis...) ?
https://quantumwise.com/documents/tutorials/latest/InAsDevice/index.html/chap.gatescan.html 

Regards_
Dipankar

[Dipankar Saha]

How to know the intrinsic w.f. of the metallic box region? / Say, I cover the entire channel region with the thin slab...and set it to "0" volt (voltage between the right and the left electrode is also set to "0"). Then, how this will affect the energy band alignment?

[Jess Wellendorff]

1) Depends on what you want to simulate. If the dielectric is supposed to be spatially confined to the (central) scattering region, then no. In the tutorial you refer to, the effect of doping is investigated, so the dielectric is part of the electrodes (meaning that the dielectric extends much longer than the nanoscale width of the device).

2) I'm not sure what you mean. What is plottet is simply 1D Projections of the electrostatic difference potential.

[Daniele Stradi]

With respect to 2): the oscillations are actually there, but the device is so big in this case that they are hard to see If you want a "planar" average, you should consider doing a macroscopic average of the electrostatic potential. See http://arxiv.org/abs/1601.04651 and references therein (in particular, Ref. 48).

With respect to the w.f. question: if you set the gate voltage to 0 Volts, that won't affect the band alignment in the system.

[Dipankar Saha]

Thank you Jess.... I understand your point..!!

[Dipankar Saha]

Dear Daniele Stradi,

Thank you for the reply...
_________________________

Are you sure that....setting gate voltage to "0" will not affect the energy band alignment at all ??!! / Because, my observation is somewhat different (where I need a certain amt. of gate voltage to nullify the band bending or, just to reproduce the same result as that had been obtained in the case of the two-port struct.)
.... Anyways, I will send you the plots ...

Best_
Dipankar