k point sampling in negf calculation

Top Page > Browsing

k point sampling in negf calculation

Date: 2024/04/29 10:33
Name: Kieran <skn_neu@hotmail.co.uk>: Dear All,

I read the manual about the NEGF calculation part for OPENMX code.

In step 1, the manual says that one needs to 'set the a-axis for the direction of electronic transport in the band structure calculation'. If the c axis is the transport direction in my system, I just need to switch the positions for a and c axis vectors in step 1 calculation. Am I correct?

If so, do I need to change the atom coordinates as well? If I do not change the atom coordinates, I do not understand how OPENMX recoginizes the whole structure?

Another question is that, the k point sampling is along the transport direction in step 1; something like '200 1 1'. The k point sampling along y or z axis should be 1. Is it correct?

Would anyone please give me some explanations?

Thank you in advance.

Kieran

Page: [1]

Re: k point sampling in negf calculation ( No.1 )

Date: 2024/04/30 15:18
Name: Kieran <skn_neu@hotmail.co.uk>

Suppose the transport in my system is along the z axis and the atom coordinates are written based on the original lattice vectors.

Now, I have to change the lattice vector in the 'Atoms.UnitVectors' section and move the z-axis lattice vector to the first line in this section for the transport calculation along the z axis; however, the mannual says that I do not need to rotate the structure. If I only change the lattice vector without changing the atom coordinates in the 'Atoms.SpeciesAndCoordinates' section, how does OpenMX recognize the structure?

I am puzzled by this issue. Would anyone please give me some explanations?

Thank you very much in advance.

Kieran

Page: [1]