Inconsistent Grid dimensions between PWSCF and Yambo

[lyzhao]

Dear all,
I run a calculation for a system with the following lattice

Code: Select all

 
 CELL_PARAMETERS {bohr}
   15.962193    0.000000    0.000000
    0.000000   13.947070    0.000000
    0.000000    0.000000    9.727789
 
 K_POINTS automatic
  3 3 5 0 0 0
 

In initialization, yambo shows the following Grid

Code: Select all

Grid dimensions      :   5   3   3

It seems to be inconsistent, what happens?
How to make them consistent？

r_setup and pwscf input files are attached.

Best regards,
Youzhao Lan

[Daniele Varsano]

Dear Youzhao,

Most probably all is correct (the QE output is not attached).
Please compare the k points listed by Yambo and the ones of QE output and check it, considering that now the K grid is
ordered as follows and the first K direction is along z.

Code: Select all

Base K vectors       :  K_min[ 1 ]  K_min[ 2 ]  K_min[ 3 ]
  K_min[ 1 ] :  0.000000  0.000000 -0.200000 [rlu]
  K_min[ 2 ] :  0.333333  0.000000  0.000000 [rlu]
  K_min[ 3 ] :  0.000000 -0.333333  0.000000 [rlu]

Best,
Daniele

[lyzhao]

Dear Daniele,
Thanks for your quick reply.
PWSCF outfile is attached.
And could you tell what the meaning of K_min is?

Best regards.

[Daniele Varsano]

Dear Youzhao,

the k points of qe and yambo are indeed the same, you can compare the list reported in the r_setup in the section

Code: Select all

[02.04] K-grid lattice 

with the list in the qe output in cryst. coord..

Yambo just ordered in different way the basis of the k vectors.
K_min is the smallest k vector present in the grid for each vector of the basis, ie essentially the spacing of the grid.
As you can see, the index 1 is along z, the second along x and the third along y, so the yambo grid (5,3,3) is exactly the same of qe (3,3,5).
Best,
Daniele