[Wien] Supercell setup

[Shu Miao]

By FAQ, I think the correct procedure to setup a supercell is as follow:
1. Using StructGen to generate a basic cell.
2. Expend the basic in x/y/z to the expected size. Hence, we have a big
cell with many atoms.
3. Make some modification in the big cell, such as atomic position, atom
type or atom label to break the symmetry.
4. Run initialization as usual. Some programs during the initialization
may make changes on the existing files due to the changes of the
"composition" or symmetry. Just let it be.
5. Now the big cell has nothing special. You can do anything you want.

Is above understanding correct?
If so, I don't know what's wrong with my calculation. The initial
case.struct file works fine to calculate properties. But the supercell
program cannot run with it. Who can help?

Thanks

On Mon, 25 Oct 2004, Shu Miao wrote:

> Dear Wien users,
> I have a very basic question concerning the supercell setup.
> My material is #166(R-3m) with O at (0,0,0.26), Mn at (0,0,0) and Li at
> (0,0,0.5). [These are hexagonal coordinate, and they are easily converted
> to (0.26,0.26,0.26), (0,0,0) and (0.5,0.5,0.5) in rhombohedral system.]
> I want to replace half Mn with Ni. So seems I need to build a supercell.
> I created the case.struct by StructGen. Then run "supercell" from the
> command line. Only specified the dimensions to double along x,y,z and kept
> other default setting. The program wouldn't run, here is the displayed
> message:
> >>>>
>  Program generates supercell from a WIEN struct file.
>
>  Filename of struct file:
>
>  Fatal Error occured:
>  Unknown lattice type: R
>
>  Program terminated.
>
> 0.001u 0.000s 0:00.00 0.0%	0+0k 0+0io 61pf+0w
> >>>>
> I thought this was because all atoms were on their ideal positions as told
> in the FAQ. I tried to modify the structure from StructGen. Putting lables
> on Li/Mn, it doesn't work (it should not work since only one position
> for Li/Mn in the cell). Displacing one of the O position slightly, this
> changed the O multiplicity from 2 to 6. But when run supercell to double
> the size, above message appeared again.
> I know there must be something wrong in my method to break the symmetry.
> But I don't know where this mistake was.
> So the question become quite basic--how to make some atomes different in
> the case. struct file or in StructGen? I worried that the multiplicity
> will also work on the modifications so that more positions thant I expect
> will be changed, even if I can make some change on the file. Who can give
> me advice to have the "supercell" run?
>
> Thanks a lot
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>