[Wien] Space group change during running x sgroup

[Gavin Abo]

View case.struct of the supercell in StructGen of w2web.  If the 
supercell has a space group, it will show what it is.  If it has no 
space group, it should instead have a general lattice highlighted such 
as "P".

The WIEN2k 19.1 usersguide [1] states on page 110 under section "6.2 
SGROUP":

/This program uses information from case.struct (lattice type, lattice 
constants, atomic positions)//
//and determines the spacegroup .../

 From the above statement, you can probably derive that a purpose of the 
sgroup program is to take a case.struct as input having a structure 
defined in a general lattice having an unknown spacegroup and gives as 
output what space group fits the structure.  Therefore, the space group 
should be displayed after clicking "x sgroup" button in w2web or should 
be found in the case.outputsgroup file.

Since Phonopy can only parse space group 1_P1 as input [2], the output 
case.struct files from it are probably general P lattice that would also 
have space group 1_P1.  Phonopy might not add the space group or special 
labels in the output case.struct files.  If so, the special labels may 
need to be added manually (based on [3]) to all atoms so that "x sgroup" 
finds space group 1_P1 else it might reduce the cell to one with higher 
symmetry.

The triclinic space group P1 [4] has the lowest symmetry were there are 
no restrictions on the values of the cell parameters [5] in which it 
should have the advantage that any structure is definable in it.  The 
disadvantage that comes from that is it computationally extensive for 
WIEN2k calculations which is probably not surprising with a higher 
number of inequivalent positions compared to if you were able to group 
atomic positions into sets of equivalent position to make use of 
symmetry [6,7]. Typically WIEN2k users are interested in comparing a 
structure with slight parameter changes [9,10] in which a change in 
basis vectors of the structure [10] would be undesirable.

So, you would need to use your crystallographic knowledge to use "x 
sgroup" as to not have GIGO [11] that you may have learned such as be 
reading the Volumes on International Tables for Crystallography [12].

Bilbao Crystallographic Server's Subgroups might also be of interest 
because at [13] it mentions it can list other compatible space groups 
for a supercell.


[1] http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf
[2] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14219.html
[3] http://susi.theochem.tuwien.ac.at/reg_user/faq/supercells.html
[4] 
https://en.wikipedia.org/wiki/Space_group#Table_of_space_groups_in_3_dimensions
[5] http://xrayweb.chem.ou.edu/notes/symmetry.html
[6] http://chemistry.bd.psu.edu/jircitano/6symmetry.pdf
[7] 
https://www.mindat.org/article.php/2721/Determining+Symmetry+of+Crystals%3A+An+Introduction+
[8] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg20318.html
[9] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg14836.html
[10] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg16765.html
[11] https://en.wikipedia.org/wiki/Garbage_in,_garbage_out
[12] https://it.iucr.org/
[13] https://www.cryst.ehu.es/cgi-bin/cryst/programs/subgrmag1_cell.pl

On 7/24/2020 6:55 PM, Peeyush Kumar Kamlesh wrote:
> Sir,
> One more question. How can I know the space group of the supercell? I 
> searched but was unable to find it.
> Thank you!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20200724/52b2a820/attachment-0001.html>