The ABINIT Users Mailing List ( CLOSED )

[Xavier Gonze <gonze@pcpm.ucl.ac.be>]

On 08 Jun 2006, at 18:57, v.sokhan@imperial.ac.uk wrote:

> Hi Abinitians,
>
> For my research I needed to calculate the bulk structure of alpha- 
> Mn (A12). It is perhaps one of the less trivial simple elemental  
> solids and according to crystal lattice structures database (http:// 
> cst-www.nrl.navy.mil/lattice/index.html) its BCC primitive cell  
> includes 29 atoms (58 in total for the cubic cell). I prepared an  
> input file and when I run it without BCC rprim array (commented  
> out), abinit recognises the structure as #215 (P-43m), but with  
> rprim included, abinit fails to recognise the structure at all  
> ("the space group has not been recognized") (I expected #217, I-43m  
> in this case). I have carefully checked the lattice vectors and am  
> pretty sure that they are correct. To pin down the problem, I found  
> a more simple substance with the same symmetry group, SiF4, and run  
> a job on it: an input file is attached below for your inspection.  
> The problem remains, i.e. abinit defines the symmetry as P-43m w/o  
> rprim, and "..not recognized" with rprim. Apparently, I am doing !
>  something wrong, but couldn't figure it out and would appreciate  
> very much any thoughts on the subject.

Somehow linked to the answer by Pierre-Matthieu, it seems that you do  
not appreciate
the difference between "xred" and "xcart" : actually, if you change  
"xred" to "xcart"
in your input file, ABINIT correctly finds #217, I-43m   . "xred" are  
expressed in the
coordinates defined by "rprim", while "xcart" is independent of "rprim".

Xavier