Author Topic: Kpoints vs DOS  (Read 3677 times)

0 Members and 1 Guest are viewing this topic.

Offline Jahanzaib

  • QuantumATK Guru
  • ****
  • Posts: 108
  • Country: gb
  • Reputation: 3
    • View Profile
Kpoints vs DOS
« on: June 26, 2023, 17:23 »
Dear Expert
I am really confused about the issue I am facing that I calculated DOS for my system with different K points (1,3,5,7 in C direction and A=B=1) and got the exactly same energy in every case.

- the way I do - I optimized my geometry in first step
- I given the path of optimized geometry, then calculated the DOS to perform convergences test.

I studied online and nothing found something more relavent to my problem - just got "The energy *will* be the same if your system only has localised states, for example a molecule in a large simulation cell".

For the reference, I have attached optimized file and then DOS with 1 kpoints. Can some one suggest me what's the issue and why I am getting the same energy?
« Last Edit: June 26, 2023, 23:41 by Jahanzaib »

Offline Jahanzaib

  • QuantumATK Guru
  • ****
  • Posts: 108
  • Country: gb
  • Reputation: 3
    • View Profile
Re: Kpoints vs DOS
« Reply #1 on: June 28, 2023, 02:59 »
I think my structure is symmetrical that may play a role in obtaining the same energy for different k-points. 
- I already optimized the geometry, is there a way I can put symmetrical=False in DOS.

or some one guide me from my previous post.

Thank you

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5589
  • Country: dk
  • Reputation: 102
    • View Profile
    • QuantumATK at Synopsys
Re: Kpoints vs DOS
« Reply #2 on: June 29, 2023, 21:55 »
The reason why your results are not dependent on the number k-points is simply that it's so large in the periodic direction. The longer the periodicity is, the fewer k-points you need, this is a general point.

You should not use this keyword for DOS.