Monkhorst pack vs gamma centred

[simly]

Hi,
I'm just exploring the kpoints here and pardon this trivial question.

Are there any significant differences between the option of using Monkhorst pack and Gamma (gamma centred grid) in the automatic mesh generation besides the shift from G for Monkhorst pack?

Which would make a better calculation for bulk?

Thanks for your help.

[alex]

Hi,

this becomes important for hexagonal cells. Non-Gamma centred k-point meshes do not reflect the symmetry accordingly (if I'm not mistaken). Check the VASP-Winterschool tutorial on that topic.

alex

[simly]

Thanks for your answer =)
Does it mean if the structure is non-hexagonal, non-gamma centred and gamme centred will not make significant difference?

[admin]

you have to test, it depends on details of the electronic structure.
In general, for low-symmetry cells it is sometimes difficult to symmetrize the k-mesh if it is not centered ON Gamma, the hexagonal cell was just one representative example chosen for the workshop.

[simly]

The manual states that the following is the Gamma centered grid. Is it true that the following can be considered as the Gamma centered MP grid? By replacing it with the "Gamma" with "Monkhorst Pack" the sole difference lies in the fact that the former is centered on gamma while the latter has a shift?

Sorry for the question. I just want to make sure this is the correct understanding of the manual. Thanks!


Automatic Mesh
0
Gamma
2 2 2
0 0 0

[job]

My understanding based on reading the manual (http://cms.mpi.univie.ac.at/vasp/vasp/node60.html ) is that the Gamma grid is indeed made with the Monkhorst-Pack scheme, but after the generation the grid is shifted so that one of the k-points lies on the gamma point.

Or in other words, for a grid with odd divisors, there is no difference between the Gamma and MP scheme. They produce exactly the same result.

So yes, your understanding is correct.