Abinit kpoint grid generation and kptrlen

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
Boris
Posts: 128
Joined: Tue Feb 16, 2010 10:13 am
Location: France

Abinit kpoint grid generation and kptrlen

Post by Boris » Sat Mar 12, 2011 12:20 am

Dear all,

For curiosity purpose, I tried to use the abinit kpoint grid generation, the one that is performed when neither ngkpt or kprtlatt are defined. And I think I don't understand how it works and what Iit is supposed to do.

My trial system is an hexagonal lattice and my convergence tests lead me to choose a 11x11x9 kpoint grid (it's a metal). With this mesh, there are 80 kpoints in the BZ and kptrlen=8.29595466E+01, which is HUGE compare to what abinit can find with its grids. Below is abinit's opinion about kpoint grids for my system:

Code: Select all

 List of best grids, ordered by nkpt.
  (stop at a value of kptrlen 20% larger than the target value).
  (the merit factor will tend to one or two in 3 dimensions)
  (and to one, two or four in 2 dimensions)

    nkpt   kptrlen    grid#  merit_factor
     1    1.3063E+01    13      0.2022
     2    1.5084E+01     3      0.1557
     3    2.6125E+01    15      0.5393
     4    2.6370E+01     5      0.4159
     8    3.0167E+01     6      0.3114
    10    3.7709E+01     7      0.4865

 For target kptrlen=    3.0000E+01, the selected grid is number     6,
     giving kptrlen=    3.0167E+01 with nkpt=     8


What do these results mean? The lowest kptrlen is obtained with a very (very) coarse grid with only 1 kpoint. This is definitely not the best grid... Even the best rated grid has only 3 kpoints.

I don't understand what we are supposed to get from these data.

Thank you for your insight

Boris
----------------------------------------------------------
Boris Dorado
Atomic Energy Commission
France
----------------------------------------------------------

mverstra
Posts: 655
Joined: Wed Aug 19, 2009 12:01 pm

Re: Abinit kpoint grid generation and kptrlen

Post by mverstra » Mon Mar 14, 2011 2:58 pm

kprtlen is the inverse of the shortest kpt length, so more is better precision.

The grid it is recommending has 8 k-points (1st column) and is numbered 6 in the order in which abinit generated the set of grids to test.

This is the first which satisfies your request that kptrlen >= 3

M.
Matthieu Verstraete
University of Liege, Belgium

Locked