[CP2K-user] What is the principle of &fixed_atoms?

[fa...@gmail.com]

The fractional coordinates describe the position with respect to the cell 
dimension while the cartesian ones reference with respect to the the origin 
(0,0,0). Therefore, changing the cell dimension (CELL_OPT) you modify the 
reference for the fractional coordiates but not for the cartesian ones.

Let me give you a simple example:

Assume we have a 1D system, with a cell from 0 to L and one atom in the 
middle. The fractional coordinate of the atom is 0.5, the cartesian one  
L/2.
Now you fix the fractional coordinate of the atom (such that the atom 
remains exactly in the middle of the cell) and change the cell length to 
2*L. The new cartesian coordinate is now (2*L)/2=L even though the atom was 
"fixed".

Best,
Fabian

On Friday, 7 August 2020 at 06:42:49 UTC+2 ka... at gmail.com wrote:

> I actually understood hardly but I think I got some hints thanks to you 
> all.
>
>
> If anyone wants to supplement the explanation, please write here.
> Again, thank you everyone.
>
>
>
> 2020년 8월 6일 목요일 오후 10시 49분 4초 UTC+9, Travis 님의 말:
>
>> Hi,
>>
>> Fractional coordinates are converted back to Cartesian by multiplying by 
>> the lattice lengths and shape (if not 90 degrees). Cartesian coordinates 
>> can then change if the box size changes.
>>
>> Through freezing or thawing atom positions, fixing angles, or fixing 
>> symmetry you can perform pretty much all the same calculations as listed on 
>> the VASP wiki for the isif keyword (has an easy to read table).
>>
>> -T
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200807/e7c531fa/attachment.htm>