[Pw_forum] ibrav7 doesn't show the structure of BaNi2As2 correctly

[Tone Kokalj]

On Wed, 2014-05-28 at 10:43 -0700, David Foster wrote:
>  Dear Prof. Kokalj
> 
> Maybe this question bored you, but it is important for me. Sorry for any inconvenience. 
> 
> Would you please give me more information that how to recognize the difference in the lattice definition. 
> If I am right in both softwares (QE and MS) A=B<C. 
> 
> So, how did you recognize the difference? 

Actually I didn't (at least initially). But from experience I know that
such kinds of problems are typically due to mismatches in lattice
definitions. Your reference to A=B<C pertains to conventional
tetragonal-body-centered cell. This typically implies that conventional
lattice vectors are 

V1=(a,0,0)
V2=(0,b,0)
V3=(0,0,c)

But then there are multiple choices how to choose primitive vectors. The
QE definition is (see INPUT_PW.txt):

v1=(a/2)(1,-1,c/a)  
v2=(a/2)(1,1,c/a) 
v3=(a/2)(-1,-1,c/a)

but it can equally well be this (1st and 3rd vectors swapped):

v1=(a/2)(-1,-1,c/a)
v2=(a/2)(1,1,c/a) 
v3=(a/2)(1,-1,c/a)

or any other possibility.  To see what effect this have when the
coordinates are specified in fractions of lattice vectors (i.e. crystal
coordinates), consider these two possibilities (patch appropriately your
input file):


&system
     ibrav=0,  celldm(1)=7.83745,
...

 CELL_PARAMETERS
  0.5 -0.5 1.400757
  0.5  0.5 1.400757
 -0.5 -0.5 1.400757

 ATOMIC_POSITIONS crystal
  Ba       0.00000   0.00000   0.00000  
  As       0.34710   0.34710   0.00000  
  As       0.65290   0.65290   0.00000  
  Ni       0.75000   0.25000   0.50000  
  Ni       0.25000   0.75000   0.50000  


versus:


&system
     ibrav=0,  celldm(1)=7.83745,
...
 CELL_PARAMETERS
 -0.5 -0.5 1.400757
  0.5  0.5 1.400757
  0.5 -0.5 1.400757

 ATOMIC_POSITIONS crystal
  Ba       0.00000   0.00000   0.00000  
  As       0.34710   0.34710   0.00000  
  As       0.65290   0.65290   0.00000  
  Ni       0.75000   0.25000   0.50000  
  Ni       0.25000   0.75000   0.50000  


The second version gives correct structure, but the first version not.


Regards, Tone


PS: David, please just call me "Tone" w/o any "prof." prefix, which I am not.
-- 
Anton Kokalj
J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia 
(tel: +386-1-477-3523 // fax:+386-1-477-3822)

Please, if possible, avoid sending me Word or PowerPoint attachments.
See:  http://www.gnu.org/philosophy/no-word-attachments.html