[Wien] Re: about the question of symmetry in antiferromagnetic CoO

[Steven Homolya]

On Fri, 5 Mar 2004 01:06 am, you wrote:
> Dear Steve,
> Thanks a lot! But be frank, I think there is some misunderstanding.

OK, I'll be Frank.

> I can not understand how the "bcc" structure come.

Not "bcc" just "bc" (i.e., not cubic). Just look at the monoclinic struct file 
with xcrysden. There is a bc structure within the unit cell that stares one 
in the face.

If you are trying to impose a magnetic unit (super)cell over your normal unit 
cell then you'll have to do so by labelling the appropriate atoms as 
inequivalent from the start.

Best regards,

Frank


> The ferromagnetic CoO is
> a FCC structure, for antiferromagnetic, in (111) face, it is still
> ferromagnetic, but for the nearest layers along (111) face, the spin are of
> opposite directions. So, the supercell is still a FCC structure with many
> equivalent atoms. I put the structure this way simply because I want to
> tell the program where the atoms are. And then I want to see what cells the
> program will generate for me. Yes, it generated a monoclinic cell with 1Co
> (up), 1 Co (dn), and 2 O. It is right. But when I accepted the structure
> file they suggested and save it. There is a signal "Error - incorrect space
> group symbol". I don't know why it appears here since it is just the file
> they suggested. I use the file to do the "nn" and "sgroup", it is also
> fine, but if I go further to do the "symmetry", I will be told "-3 atom
> missing".
>
> I don't know whether I made it clear. If I give the correct positions for
> every atom, the programs should be able to give me the optimal structure
> file. But if I accept this structure file, I can not get right rotation
> matrix in symmetry.
>
> Thanks a million again for your interest!
>
> Nice day!
>
> Sincere
> Xinzheng
>
> -----Original Message-----
> From: Steven Homolya [mailto:steven.homolya at spme.monash.edu.au]
> Sent: Thursday, March 04, 2004 1:03 PM
> To: xinzheng
> Cc: wien at zeus.theochem.tuwien.ac.at
> Subject: Re: about the question of symmetry in antiferromagnetic CoO
>
> DX,
>
> I looked at your struct files, and it seems that you have a nice little
> body-centred tetragonal unit cell with two inequivalent atoms. The 64 atom
> struct file is just plain silly, unless you want to replace one of the
> atoms
>
> with a defect/vacancy. The monoclinic structure is also unnecessary, since
> you obviously have tetragonal symmetry there.
>
> To get a proper struct file just put the 2 (4) atoms where they are
> supposed
>
> to go in a tetragonal lattice, using w2web with the correct spacegroup (or
> using a hand-made struct file).
>
> S
>
> PS. I did not check your struct files with wien2k. Note that you need high
> numerical precision in your struct file for symmetry/sgroup to work.
>

-- 
Steven Homolya
School of Physics and Materials Engineering
Monash University, VIC 3800
Australia
Tel: +61 3 9905 3694
Fax: +61 3 9905 3637