Bulk Si CBS

[ramkrishna]

Dear Sir,
       Can you please tell me how to calculate complex band structure of BULK Silicon in different crystallographic direction [i.e. 100, 110, 111] ? In the tutorial http://www.quantumwise.com/publications/tutorials/mini-tutorials/132-complex-band-structure , Si [100] surface is considered. Is it actually for bulk silicon calculation or nanowire? If it is for bulk, then how can I specify the complex band structure from conduction band minima to valence band maxima  for bulk?

Regaards
Ramkrishna

[Anders Blom]

Unfortunately you have found the CBS tutorial via Google, and not via a link on our website - it is not a ready tutorial yet, we've been postponing it awaiting some new features in ATK 12.2 which will make it much easier to set up the structure, etc.

If you can figure something out anyway you're of course welcome to use it The complex band structure applies to a crystal direction, so it's in the bulk but for a specific direction. Thus it's not a surface in the sense that there is vacuum on top, but you do need to have a well-defined direction which would correspond to a surface plane.

[ramkrishna]

Dear SIr,
         I am trying to calculate CBS of bulk silicon for different crystallographic directions [i.e. 100, 110, 111] and want to reproduce the complex band structure as given in these articles http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5091095&tag=1 and http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5530371&tag=1 . In this article the authors have used sp3d5s* method including spin-orbit coupling. Now, when I am trying to calculate by using a script which is given in  http://quantumwise.com/documents/manuals/latest/ReferenceManual/XHTML/ref.complexbandstructure.html, the band gap value is about 1.17eV for 100 direction and the complex band structure nature is nearly same as reported in those articles but when I am trying to calculate in other directions, it gives higher band gap value (around 1.5eV). In that article, the band gap value for all directions is reported as 1.12eV. I can't understand whether my calculations are correct or not. If it a wrong procedure, then please suggest me some approach. It will be too much helpful for me.

Some few points I also want to know:
1. Is it not possible at all to calculate CBS for indirect band gap semiconductor as it is required phonon ?
2. Is there any plan to implement spin-orbit coupling and phonon interaction in quantumwise ?

Regards
Ramkrishna

[Anders Blom]

I haven't had opportunity to try the different crystal directions myself, so let me limit my answer to your two questions:

1. The nature of the band gap has no real influence on the CBS calculation, but as you realize the CBS is always computed at a single k-point (kx,ky).
2. We do plan to implement spin-orbit interaction, although it's not at the top of the priority list before the summer. As for phonons, it's a rather difficult matter; our plan is to first just be able to compute the phonon spectrum.

[ramkrishna]

Dear Sir,
        

The nature of the band gap has no real influence on the CBS calculation, but as you realize the CBS is always computed at a single k-point (kx,ky).

So, I have to fix kx and ky. Now, for this case I am little bit confused, if I have got a band structure for [G, Z] direction where the G is (0,0,0) and Z is (0,0,1/2) and the conduction band minima is within G,Z (lets say 0,0,0.25 ) and valence band maxima is at G (I mean, indirect band gap), then if I put kx= 0 , ky= 0, can I say that the CBS is within conduction band minima and valence band maxima, i.e. within indirect band gap? or is this CBS only at the G point (direct band gap)? And one more thing, how can I differ a CBS for direct and indirect band gap semiconductor for this type of case?

Regards
Ramkrishna

[ramkrishna]

Dear Sir,
        In case of layered ribbon structure, for 1Layer, 2Layer the CBS option is coming in script generator (analysis) but for 3Layer it is not coming, so is it not possible to calculate CBS for 3Layer system?? If it is possible, then what shall I do for that??

Regards
Ramkrishna

[Anders Blom]

To compute the CBS it is necessary that the system fulfills the conditions of an electrode, i.e. that the AB vectors are both perpendicular to C.

[Anders Blom]

In ATK 12.8 we have updated the complex band structure (cbs) calculations a lot. It is for instance no longer necessary to set the system up as an electrode, and we have also improved the computational algorithm quite a bit.

For more details see the new tutorial, available via http://quantumwise.com/tutorials

NOTE: There are bugs in the cbs functionality for 12.8.b1 and b2. These are fixed for the final release, but please don't use the beta-versions for production runs.