Dear all,
I have been trying to plot the projected bulk band structure(PBBS) to extract surface states.
I calculated Surface band structure by using slab model.
And I know the basic principle of PBBS.
But I don't know the details of calculating and plotting the PBBS, such as how to project the band along the K_z to the plane of K_//(x,y).
Could anyone help me with this, please? Any tips are highly appreciated.
Best regards,
Zhe
how to calculate and plot Projected Bulk Band Structure
Moderator: bguster
-
psychologist
- Posts: 1
- Joined: Mon Apr 25, 2011 11:25 am
-
sevket simsek
- Posts: 18
- Joined: Tue May 11, 2010 1:05 pm
Re: how to calculate and plot Projected Bulk Band Structure
Dear psychologist
I try to calculate surface band structure of materials. ı don't know how it is calculated.
can you send me your input file for surface band structure.
I try to calculate surface band structure of materials. ı don't know how it is calculated.
can you send me your input file for surface band structure.
Re: how to calculate and plot Projected Bulk Band Structure
Hi,
I don't know an *automated* way to do this in any DFT code, but you can do this from first principles with some effort. First use your surface slab model to calculate a band structure around a surface BZ path (should be 2D). Then, using the bulk structure for your material (ie, a small primitive cell), you need to run many conventional BS calculations around a path that is equivalent to your surface BZ path but offset along k_perp by an increasing value.
The difficulty is translating the path from the surface BZ back to the equivalent path in the bulk BZ. Once you have done that, just take that same set of k-points and add a constant k_{perp} for increasing values of k_{perp} until you reach the BZ boundary.
There are further subtleties if the path in the bulk starts to intersect the BZ boundaries for some intermediate value of k_{perp}.
In any case, if you do this for a sufficient number of k_{perp} values and plot the bandstructures overlaying each other it is easy to see which regions are forbidden and so on - you can shade the projected band structure however you like (again, no automated way of doing this) and then overlay the actual surface slab band structure to see which states are resonances and so on.
I have only done this in CASTEP and not Abinit so I don't have an example to send you but if you understand the physics of what is going on it should be straightforward to follow what I'm saying here.
Kane
I don't know an *automated* way to do this in any DFT code, but you can do this from first principles with some effort. First use your surface slab model to calculate a band structure around a surface BZ path (should be 2D). Then, using the bulk structure for your material (ie, a small primitive cell), you need to run many conventional BS calculations around a path that is equivalent to your surface BZ path but offset along k_perp by an increasing value.
The difficulty is translating the path from the surface BZ back to the equivalent path in the bulk BZ. Once you have done that, just take that same set of k-points and add a constant k_{perp} for increasing values of k_{perp} until you reach the BZ boundary.
There are further subtleties if the path in the bulk starts to intersect the BZ boundaries for some intermediate value of k_{perp}.
In any case, if you do this for a sufficient number of k_{perp} values and plot the bandstructures overlaying each other it is easy to see which regions are forbidden and so on - you can shade the projected band structure however you like (again, no automated way of doing this) and then overlay the actual surface slab band structure to see which states are resonances and so on.
I have only done this in CASTEP and not Abinit so I don't have an example to send you but if you understand the physics of what is going on it should be straightforward to follow what I'm saying here.
Kane
Dr Kane O'Donnell
Postdoctoral Research Fellow
Australian Synchrotron
Postdoctoral Research Fellow
Australian Synchrotron